CN111476117A - Safety helmet wearing detection method and device and terminal - Google Patents
Safety helmet wearing detection method and device and terminalDownload PDFInfo
- Publication number
- CN111476117A CN111476117ACN202010220020.5ACN202010220020ACN111476117ACN 111476117 ACN111476117 ACN 111476117ACN 202010220020 ACN202010220020 ACN 202010220020ACN 111476117 ACN111476117 ACN 111476117A
- Authority
- CN
- China
- Prior art keywords
- image
- helmet wearing
- detected
- safety helmet
- wearing detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
Translated fromChinese
Description
Translated fromChinese技术领域technical field
本申请属于安全管理技术领域，尤其涉及一种安全帽佩戴检测方法、装置和终端。The present application belongs to the technical field of safety management, and in particular, relates to a safety helmet wearing detection method, device and terminal.
背景技术Background technique
安全帽是指对人头部受坠落物及其他特定因素引起的伤害起防护作用的帽子。在许多作业环境中，施工方会要求工人在作业时佩戴好安全帽，安全帽的佩戴检测也成为施工方安全管理的重要一环。A safety helmet refers to a hat that protects the human head from injuries caused by falling objects and other specific factors. In many operating environments, the construction party will require workers to wear safety helmets during operation, and the wearing detection of safety helmets has also become an important part of the construction party's safety management.
现有的安全帽佩戴检测方法往往是通过将待检测图像输入到安全帽佩戴检测模型中完成安全帽佩戴检测，然而，这种方法精度低，很难准确识别出在待检测图像中面积占比较小的目标是否佩戴了安全帽，不能有效保障工人在作业时的安全。The existing helmet wearing detection methods often complete the helmet wearing detection by inputting the image to be detected into the helmet wearing detection model. However, this method has low precision and is difficult to accurately identify the area in the image to be detected. Whether the small target is wearing a safety helmet cannot effectively ensure the safety of workers during operation.
发明内容SUMMARY OF THE INVENTION
本申请实施例提供一种安全帽佩戴检测方法、装置、终端和计算机可读存储介质，可以提高安全帽佩戴检测的精度。Embodiments of the present application provide a method, device, terminal and computer-readable storage medium for safety helmet wearing detection, which can improve the accuracy of safety helmet wearing detection.
本申请实施例第一方面提供一种安全帽佩戴检测方法，包括：A first aspect of the embodiments of the present application provides a safety helmet wearing detection method, including:
获取待检测图像；Obtain the image to be detected;
对所述待检测图像进行分割处理，得到多张待检测子图像；Performing segmentation processing on the to-be-detected image to obtain a plurality of to-be-detected sub-images;
将所述待检测图像和所述待检测子图像输入预先建立的安全帽佩戴检测模型，由所述安全帽佩戴检测模型对所述待检测图像和所述待检测子图像进行安全帽佩戴检测，并输出携带安全帽佩戴标记信息的标记图像和携带安全帽佩戴标记信息的标记子图像；Inputting the to-be-detected image and the to-be-detected sub-images into a pre-established helmet wearing detection model, and the helmet-wearing detection model performs helmet-wearing detection on the to-be-detected image and the to-be-detected sub-images, And output the marked image carrying the safety helmet wearing marking information and the marking sub-image carrying the safety helmet wearing marking information;
分别将所述标记图像和所述标记子图像携带的安全帽佩戴标记信息映射到所述待检测图像，得到所述待检测图像对应的携带安全帽佩戴标记信息的第一目标图像。The safety helmet wearing marking information carried by the marking image and the marking sub-image is respectively mapped to the to-be-detected image, and a first target image corresponding to the to-be-detected image carrying the safety helmet wearing marking information is obtained.
本申请实施例第二方面提供的一种安全帽佩戴检测装置，包括：A safety helmet wearing detection device provided by the second aspect of the embodiment of the present application includes:
获取单元，用于获取待检测图像；an acquisition unit for acquiring an image to be detected;
分割单元，用于对所述待检测图像进行分割处理，得到多张待检测子图像；a segmentation unit, configured to perform segmentation processing on the to-be-detected image to obtain a plurality of to-be-detected sub-images;
检测单元，用于将所述待检测图像和所述待检测子图像输入预先建立的安全帽佩戴检测模型，由所述安全帽佩戴检测模型对所述待检测图像和所述待检测子图像进行安全帽佩戴检测，并输出携带安全帽佩戴标记信息的标记图像和携带安全帽佩戴标记信息的标记子图像；The detection unit is used to input the image to be detected and the sub-image to be detected into a pre-established helmet wearing detection model, and the image to be detected and the sub-image to be detected are performed by the helmet wearing detection model. Helmet wearing detection, and output the marked image carrying the safety helmet wearing marking information and the marking sub-image carrying the safety helmet wearing marking information;
映射单元，用于分别将所述标记图像和所述标记子图像携带的安全帽佩戴标记信息映射到所述待检测图像，得到所述待检测图像对应的携带安全帽佩戴标记信息的第一目标图像。a mapping unit, configured to map the helmet wearing marking information carried by the marked image and the marked sub-image to the to-be-detected image, respectively, to obtain a first target carrying the helmet-wearing marking information corresponding to the to-be-detected image image.
本申请实施例第三方面提供一种终端，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述方法的步骤。A third aspect of an embodiment of the present application provides a terminal, including a memory, a processor, and a computer program stored in the memory and running on the processor, where the processor implements the above method when executing the computer program A step of.
本申请实施例第四方面提供一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现上述方法的步骤。A fourth aspect of the embodiments of the present application provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the foregoing method are implemented.
第五方面，本申请实施例提供了一种计算机程序产品，当计算机程序产品在终端设备上运行时，使得终端设备执行时实现方法的步骤。In a fifth aspect, an embodiment of the present application provides a computer program product, which, when the computer program product runs on a terminal device, enables the terminal device to execute the steps of the method.
本申请实施例中，通过对待检测图像进行分割处理，得到多张待检测子图像，并将待检测图像和待检测子图像输入预先建立的安全帽佩戴检测模型中，不仅通过安全帽佩戴检测模型对待检测图像进行安全帽佩戴检测，使得终端能够检测到在待检测图像中面积占比较大的目标是否佩戴了安全帽，而且由于在待检测图像中面积占比较小的目标在待检测子图像中的面积占比更大，使得终端也能够通过安全帽佩戴检测模型对待检测子图像进行的安全帽佩戴检测，检测到在待检测图像中面积占比较小的目标是否佩戴了安全帽，因此，在分别将安全帽佩戴检测模型输出的标记图像和安全帽佩戴检测模型输出的标记子图像携带的安全帽佩戴标记信息映射到待检测图像后得到的第一目标图像中，不仅包含在待检测图像中面积占比较大的目标对应的安全帽佩戴标记信息，而且包含在待检测图像中面积占比较小的目标对应的安全帽佩戴标记信息，提高了安全帽佩戴检测的精度。In the embodiment of the present application, by dividing the image to be detected, a plurality of sub-images to be detected are obtained, and the images to be detected and the sub-images to be detected are input into the pre-established safety helmet wearing detection model, not only through the safety helmet wearing detection model The helmet wearing detection is performed on the image to be detected, so that the terminal can detect whether the target with a larger area in the image to be detected is wearing a helmet, and because the target with a smaller area in the image to be detected is in the sub-image to be detected. The area of the image to be detected is larger, so that the terminal can also detect the helmet wearing of the sub-image to be detected through the helmet wearing detection model, and detect whether the target with a smaller area in the image to be detected is wearing a helmet. The marking image output by the helmet wearing detection model and the helmet wearing marking information carried by the marking sub-image output by the helmet wearing detection model are respectively mapped to the first target image obtained after the image to be detected, not only included in the image to be detected The helmet wearing mark information corresponding to the target with a larger area ratio is included, and the helmet wearing mark information corresponding to the target with a small area ratio in the image to be detected is included, which improves the accuracy of the helmet wearing detection.
附图说明Description of drawings
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only for the present application. In some embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.
图1是本申请实施例提供的一种安全帽佩戴检测方法的实现流程示意图；Fig. 1 is a schematic diagram of the implementation flow of a safety helmet wearing detection method provided by an embodiment of the present application;
图2是本申请实施例提供的对待训练的安全帽佩戴检测模型进行训练的实现流程示意图；2 is a schematic diagram of an implementation flow for training a helmet wearing detection model to be trained provided by an embodiment of the present application;
图3是本申请实施例提供的安全帽佩戴检测示意图；3 is a schematic diagram of a safety helmet wearing detection provided by an embodiment of the present application;
图4是本申请实施例提供的一种安全帽佩戴检测装置的结构示意图；4 is a schematic structural diagram of a safety helmet wearing detection device provided by an embodiment of the present application;
图5是本申请实施例提供的终端的结构示意图。FIG. 5 is a schematic structural diagram of a terminal provided by an embodiment of the present application.
具体实施方式Detailed ways
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。基于本申请的实施例，本领域技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例，都属于本申请保护的范围。In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application will be described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.
安全帽是指对人头部受坠落物及其他特定因素引起的伤害起防护作用的帽子。在许多作业环境中，施工方会要求工人在作业时佩戴好安全帽，安全帽的佩戴检测也成为施工方安全管理的重要一环。A safety helmet refers to a hat that protects the human head from injuries caused by falling objects and other specific factors. In many operating environments, the construction party will require workers to wear safety helmets during operation, and the wearing detection of safety helmets has also become an important part of the construction party's safety management.
现有的安全帽佩戴检测方法往往是通过将待检测图像输入到安全帽佩戴检测模型中完成安全帽佩戴检测，然而，这种方法精度低，很难准确识别出在待检测图像中面积占比较小的目标是否佩戴了安全帽，不能有效保障工人在作业时的安全。The existing helmet wearing detection methods often complete the helmet wearing detection by inputting the image to be detected into the helmet wearing detection model. However, this method has low precision and is difficult to accurately identify the area in the image to be detected. Whether the small target is wearing a safety helmet cannot effectively ensure the safety of workers during operation.
基于此，本申请实施例提供一种安全帽佩戴检测方法、装置、终端和计算机可读存储介质，可以提高安全帽佩戴检测的精度。Based on this, the embodiments of the present application provide a safety helmet wearing detection method, device, terminal and computer-readable storage medium, which can improve the safety helmet wearing detection accuracy.
为了说明本申请的技术方案，下面通过具体实施例来进行说明。In order to illustrate the technical solutions of the present application, the following specific embodiments are used for description.
图1示出了本申请实施例提供的一种安全帽佩戴检测方法的实现流程示意图，该方法可以应用于终端，可以由终端上配置的安全帽佩戴检测装置执行，适用于需提高安全帽佩戴检测精度的情形。FIG. 1 shows a schematic diagram of the implementation process of a safety helmet wearing detection method provided by an embodiment of the present application. The method can be applied to a terminal and can be executed by a safety helmet wearing detection device configured on the terminal, and is suitable for improving the safety helmet wearing. detection accuracy.
上述安全帽佩戴检测方法可以包括步骤101至步骤104。The above-mentioned safety helmet wearing detection method may include
步骤101，获取待检测图像。
其中，上述待检测图像表示需要进行安全帽佩戴检测的图像。一般地，终端可以获取由安装在施工现场的监控设备拍摄的监控视频，并将所述监控视频中的每一帧监控图像作为待检测图像进行安全帽佩戴检测。Wherein, the above-mentioned image to be detected represents an image that needs to be detected for wearing a helmet. Generally, the terminal can acquire the surveillance video captured by the surveillance equipment installed on the construction site, and use each frame of the surveillance image in the surveillance video as an image to be detected to perform helmet wearing detection.
步骤102，对待检测图像进行分割处理，得到多张待检测子图像。In
在本申请的一些实施方式中，可以对上述待检测图像进行均等分割处理，得到多张待检测子图像。In some embodiments of the present application, the above-mentioned to-be-detected images may be equally divided to obtain a plurality of to-be-detected sub-images.
例如，在获取到像素值为1920*1080(1080P)的待检测图像之后，可以将该待检测图像均等分割为12张像素值为320*540的待检测子图像。For example, after acquiring the to-be-detected image with a pixel value of 1920*1080 (1080P), the to-be-detected image may be equally divided into 12 to-be-detected sub-images with a pixel value of 320*540.
本申请的实施例中，通过利用上述待检测图像及对上述待检测图像进行分割处理后得到的多张待检测子图像一起进行安全帽佩戴检测，不仅保证了在待检测图像中面积占比较大的目标不会因为分割处理导致该目标是否佩戴了安全帽不能被检测到，而且由于在待检测图像中面积占比较小的目标在待检测子图像中的面积占比更大，因此，可以检测到在待检测图像中面积占比较小的目标是否佩戴了安全帽。In the embodiment of the present application, by using the above-mentioned image to be detected and a plurality of sub-images to be detected obtained by dividing the above-mentioned image to be detected together to perform helmet wearing detection, it not only ensures that the area of the to-be-detected image occupies a larger proportion The target will not be detected whether the target is wearing a helmet due to segmentation processing, and because the target with a smaller area in the image to be detected occupies a larger area in the sub-image to be detected, it can be detected. Whether the target with a small area in the image to be detected is wearing a helmet.
步骤103，将待检测图像和待检测子图像输入预先建立的安全帽佩戴检测模型，由安全帽佩戴检测模型对待检测图像和待检测子图像进行安全帽佩戴检测，并输出携带安全帽佩戴标记信息的标记图像和携带安全帽佩戴标记信息的标记子图像。Step 103: Input the image to be detected and the sub-image to be detected into the pre-established safety helmet wearing detection model, and the safety helmet wearing detection model performs safety helmet wearing detection on the image to be detected and the sub-image to be detected, and outputs the marking information of carrying the safety helmet. The marker image and the marker sub-image carrying the helmet wearing marker information.
其中，上述预先建立的安全帽佩戴检测模型可以为基于单次多框检测器(SingleShot MultiBox Detector，SSD)的模型，也可以为基于卷积神经网络(ConvolutionalNeural Networks，CNN)的模型或者基于其他深度学习算法的模型。The above-mentioned pre-established helmet wearing detection model may be a model based on a Single Shot MultiBox Detector (SSD), a model based on a Convolutional Neural Networks (CNN), or a model based on other depths Models for learning algorithms.
在本申请的一些实施方式中，在将待检测图像和待检测子图像输入预先建立的安全帽佩戴检测模型之前，需要对待训练的安全帽佩戴检测模型进行训练，得到预先建立的安全帽佩戴检测模型。In some embodiments of the present application, before the image to be detected and the sub-images to be detected are input into the pre-established helmet wearing detection model, the to-be-trained helmet wearing detection model needs to be trained to obtain the pre-established helmet wearing detection model. Model.
具体的，如图2所示，上述待训练的安全帽佩戴检测模型的训练，可以包括：步骤201至步骤203。Specifically, as shown in FIG. 2 , the training of the above-mentioned safety helmet wearing detection model to be trained may include
步骤201，获取多个样本图像，以及分别与每个样本图像对应的携带预先标记的安全帽佩戴标记信息的标准图像。
在本申请的一些实施方式中，上述样本图像同样可以通过获取由安装在施工现场的监控设备拍摄的监控视频，并将所述监控视频中的每一帧监控图像作为样本图像，也可以通过获取网络上公开的包含佩戴安全帽的人员的图片，并将该图片作为样本图像。通过对上述多个样本图像进行安全帽佩戴标记信息的标记，可以得到分别与每个样本图像对应的标准图像。In some embodiments of the present application, the above-mentioned sample images can also be obtained by acquiring surveillance videos captured by monitoring equipment installed at the construction site, and using each frame of surveillance images in the surveillance video as a sample image, or by acquiring A picture of a person wearing a hard hat published on the Internet is used as a sample image. By marking the above-mentioned multiple sample images with the helmet wearing marking information, a standard image corresponding to each sample image can be obtained.
例如，可以利用LabelImg软件对上述多个样本图像进行打框标记，得到分别与每个样本图像对应的携带预先标记的安全帽佩戴标记信息的标准图像。For example, the LabelImg software can be used to frame and mark the above-mentioned multiple sample images, so as to obtain standard images corresponding to each sample image and carrying pre-marked safety helmet wearing marking information.
其中，上述安全帽佩戴标记信息至少包括未佩戴安全帽标记信息；也就是说，上述安全帽佩戴标记信息可以只包括未佩戴安全帽标记信息，也可以包括未佩戴安全帽标记信息及已佩戴安全帽标记信息；该已佩戴安全帽标记信息可以为对已佩戴红色、黄色、蓝色、白色和/或其他颜色的安全帽进行标记的标记信息。Wherein, the above-mentioned safety helmet wearing marking information at least includes not wearing safety hat marking information; that is, the above safety helmet wearing marking information may only include not wearing safety hat marking information, and may also include not wearing safety hat marking information and wearing safety helmet marking information Hat marking information; the wearing safety helmet marking information may be marking information marking red, yellow, blue, white and/or other colored safety helmets.
步骤202，将多个样本图像中的目标样本图像输入待训练的安全帽佩戴检测模型中，由待训练的安全帽佩戴检测模型输出目标样本图像对应的携带安全帽佩戴标记信息的待确认图像。Step 202: Input the target sample images from the multiple sample images into the helmet wearing detection model to be trained, and the helmet wearing detection model to be trained outputs the to-be-confirmed image corresponding to the target sample image carrying the helmet wearing label information.
其中，上述目标样本图像是指多个样本图像中的任意一个样本图像。本申请实施例中，通过利用大量的样本图像依次对待训练的安全帽佩戴检测模型进行训练，使得得到的预先建立的安全帽佩戴检测模型能够对包含不同目标的图像进行安全帽佩戴识别。Wherein, the above-mentioned target sample image refers to any one sample image among the plurality of sample images. In the embodiment of the present application, by using a large number of sample images to sequentially train the helmet wearing detection model to be trained, the obtained pre-established helmet wearing detection model can perform helmet wearing recognition on images containing different targets.
同样的，上述待确认图像中携带的安全帽佩戴标记信息也至少包括未佩戴安全帽标记信息。Similarly, the safety helmet wearing marking information carried in the above image to be confirmed also includes at least the marking information of not wearing a safety helmet.
步骤203，计算待确认图像中的安全帽佩戴标记信息与目标样本图像对应的标准图像携带的预先标记的安全帽佩戴标记信息之间的匹配度，若匹配度小于匹配度阈值，则调整待训练的安全帽佩戴检测模型的参数，重新利用目标样本图像对待训练的安全帽佩戴检测模型进行训练，直至利用目标样本图像对待训练的安全帽佩戴检测模型进行训练的次数大于或等于第一次数阈值时，或者，待确认图像中的安全帽佩戴标记信息与目标样本图像对应的标准图像携带的预先标记的安全帽佩戴标记信息之间的匹配度大于或等于匹配度阈值时，利用多个样本图像中的下一个目标样本图像对待训练的安全帽佩戴检测模型进行训练，直至待训练的安全帽佩戴检测模型的总训练次数大于或等于第二次数阈值，得到预先建立的安全帽佩戴检测模型。Step 203: Calculate the matching degree between the safety helmet wearing marking information in the image to be confirmed and the pre-marked safety helmet wearing marking information carried in the standard image corresponding to the target sample image. If the matching degree is less than the matching degree threshold, adjust the matching degree to be trained. the parameters of the helmet wearing detection model, re-use the target sample image to train the helmet wearing detection model to be trained, until the number of training times of the helmet wearing detection model to be trained using the target sample image is greater than or equal to the first time threshold When the matching degree between the safety helmet wearing marking information in the image to be confirmed and the pre-marked safety helmet wearing marking information carried by the standard image corresponding to the target sample image is greater than or equal to the matching degree threshold, multiple sample images are used. The next target sample image in the helmet wearing detection model to be trained is trained until the total number of training times of the helmet wearing detection model to be trained is greater than or equal to the second threshold, and a pre-established helmet wearing detection model is obtained.
例如，终端在获取1000个样本图像后，可以将其中的任一样本图像输入到待训练的安全帽佩戴检测模型，由待训练的安全帽佩戴检测模型输出该样本图像对应的携带安全帽佩戴标记信息的待确认图像，此时，通过计算待确认图像中的安全帽佩戴标记信息与目标样本图像对应的标准图像携带的预先标记的安全帽佩戴标记信息之间的匹配度，若该匹配度大于匹配度阈值，则利用下一个样本图像对待训练的安全帽佩戴检测模型进行训练，直至待训练的安全帽佩戴检测模型的总训练次数大于或等于第二次数阈值，得到预先建立的安全帽佩戴检测模型。For example, after acquiring 1000 sample images, the terminal can input any of the sample images into the helmet wearing detection model to be trained, and the helmet wearing detection model to be trained outputs the helmet wearing mark corresponding to the sample image The image of the information to be confirmed, at this time, by calculating the matching degree between the safety helmet wearing marking information in the image to be confirmed and the pre-marked safety helmet wearing marking information carried in the standard image corresponding to the target sample image, if the matching degree is greater than If the matching degree threshold is set, the next sample image is used to train the helmet wearing detection model to be trained, until the total number of training times of the helmet wearing detection model to be trained is greater than or equal to the second threshold, and the pre-established helmet wearing detection model is obtained. Model.
其中，上述待确认图像中的安全帽佩戴标记信息与目标样本图像对应的标准图像携带的预先标记的安全帽佩戴标记信息之间的匹配度，可以通过计算待确认图像中安全帽佩戴标记框与目标样本图像对应的标准图像中携带的预先标记的安全帽佩戴标记框之间的重合度进行计算。Wherein, the matching degree between the safety helmet wearing marking information in the image to be confirmed and the pre-marked safety helmet wearing marking information carried in the standard image corresponding to the target sample image can be calculated by calculating the safety helmet wearing marking frame in the image to be confirmed and The degree of coincidence between the pre-marked safety helmet wearing marking frames carried in the standard image corresponding to the target sample image is calculated.
需要说明的是，在本申请的一些实施方式中，在将多个样本图像中的目标样本图像输入待训练的安全帽佩戴检测模型之前，可以对样本图像进行分组，得到多组样本图像，并依次将多组样本图像中的一组样本图像输入待训练的安全帽佩戴检测模型中，对待训练的安全帽佩戴检测模型进行训练，得到预先建立的安全帽佩戴检测模型。It should be noted that, in some embodiments of the present application, before inputting the target sample images in the multiple sample images into the helmet wearing detection model to be trained, the sample images may be grouped to obtain multiple sets of sample images, and A set of sample images in the multiple sets of sample images are sequentially input into the helmet wearing detection model to be trained, and the helmet wearing detection model to be trained is trained to obtain a pre-established helmet wearing detection model.
为了加快预先建立的安全帽佩戴检测模型的运算速度，保证安全帽佩戴检测的实时性，在本申请的一些实施方式中，可以对预先建立的安全帽佩戴检测模型中的参数进行量化处理，得到量化处理后的安全帽佩戴检测模型。In order to speed up the calculation speed of the pre-established safety helmet wearing detection model and ensure the real-time performance of the safety helmet wearing detection, in some embodiments of the present application, the parameters in the pre-established safety helmet wearing detection model can be quantified to obtain Quantified helmet wearing detection model.
例如，可以利用TensorRT框架，将预先建立的安全帽佩戴检测模型中的32位浮点型参数转换为8位整型参数，得到量化处理后的安全帽佩戴检测模型，进而降低预先建立的安全帽佩戴检测模型的计算量，加快预先建立的安全帽佩戴检测模型的运算速度，提高安全帽佩戴检测的实时性。For example, the TensorRT framework can be used to convert the 32-bit floating-point parameters in the pre-established helmet wearing detection model into 8-bit integer parameters to obtain a quantized helmet wearing detection model, thereby reducing the pre-established helmet wearing detection model. The calculation amount of the wearing detection model can speed up the calculation speed of the pre-established helmet wearing detection model, and improve the real-time performance of the helmet wearing detection.
本申请的实施方式中，通过对待训练的安全帽佩戴检测模型进行训练，得到预先建立的安全帽佩戴检测模型，并将待检测图像和待检测子图像输入预先建立的安全帽佩戴检测模型，由安全帽佩戴检测模型对待检测图像和待检测子图像进行安全帽佩戴检测，可以得到携带安全帽佩戴标记信息的标记图像和携带安全帽佩戴标记信息的标记子图像。In the embodiment of the present application, a pre-established helmet-wearing detection model is obtained by training the helmet-wearing detection model to be trained, and the images to be detected and sub-images to be detected are input into the pre-established helmet-wearing detection model. The helmet wearing detection model performs the helmet wearing detection on the to-be-detected image and the sub-images to be detected, and can obtain a marked image carrying the helmet-wearing marking information and a marked sub-image carrying the helmet-wearing marking information.
相应的，在上述携带安全帽佩戴标记信息的标记图像和携带安全帽佩戴标记信息的标记子图像中的安全帽佩戴标记信息，与待训练的安全帽佩戴检测模型的训练过程中标准图像携带的安全帽佩戴标记信息对应，至少包括未佩戴安全帽标记信息。Correspondingly, the safety helmet wearing marking information in the marked image carrying the safety helmet wearing marking information and the marking sub-image carrying the safety helmet wearing marking information is the same as that carried by the standard image during the training process of the safety helmet wearing detection model to be trained. The safety helmet wearing marking information corresponds to at least the marking information of not wearing a safety helmet.
需要说明的是，由于上述预先建立的安全帽佩戴检测模型往往会对输入的图像的像素值大小进行限制，因此，终端在将待检测图像和待检测子图像输入到预先建立的安全帽佩戴检测模型之前，可以对待检测图像和待检测子图像分别进行压缩处理，得到预设分辨率的待检测压缩图像和待检测压缩子图像，并将待检测压缩图像和待检测压缩子图像输入预先建立的安全帽佩戴检测模型，使得待检测压缩图像和待检测压缩子图像符合上述预先建立的安全帽佩戴检测模型对输入的图像的像素大小要求。It should be noted that, because the above-mentioned pre-established safety helmet wearing detection model often limits the pixel value of the input image, the terminal is inputting the image to be detected and the sub-image to be detected into the pre-established safety helmet wearing detection model. Before the model, the to-be-detected image and the to-be-detected sub-image can be compressed respectively to obtain the to-be-detected compressed image and the to-be-detected compressed sub-image with a preset resolution, and the to-be-detected compressed image and the to-be-detected compressed sub-image are input into the pre-established The helmet wearing detection model makes the compressed image to be detected and the compressed sub-image to be detected meet the pixel size requirement of the input image by the pre-established helmet wearing detection model.
例如，在将待检测图像均等分割为12张像素值为320*540的待检测子图像之后，可以将该待检测图像及12张像素值为320*540的待检测子图像分别进行压缩处理，得到像素值为300*300的待检测压缩图像和12张像素值为300*300的待检测压缩子图像。For example, after the to-be-detected image is equally divided into 12 to-be-detected sub-images with a pixel value of 320*540, the to-be-detected image and the 12 to-be-detected sub-images with a pixel value of 320*540 can be compressed respectively, Obtain the compressed image to be detected with a pixel value of 300*300 and 12 compressed sub-images to be detected with a pixel value of 300*300.
在本申请的一些实施方式中，若对待检测图像进行分割处理得到的多张待检测子图像满足预先建立的安全帽佩戴检测模型对输入的图像的像素值大小的要求，则可以只对上述待检测图像进行压缩处理。In some embodiments of the present application, if a plurality of sub-images to be detected obtained by dividing the image to be detected meet the requirements of the pre-established helmet wearing detection model for the pixel value of the input image, then only the above-mentioned sub-images to be detected can be processed. The detected image is compressed.
步骤104，分别将标记图像和标记子图像携带的安全帽佩戴标记信息映射到待检测图像，得到待检测图像对应的携带安全帽佩戴标记信息的第一目标图像。
例如，如图3所示，在获取待检测图像31之后，可以对该待检测图像进行分割处理，得到多张待检测子图像，并对待检测图像和多张待检测子图像分别进行压缩处理，得到待检测压缩图像305和待检测压缩子图像301、302、303和304；此时，可以将待检测压缩图像305和待检测压缩子图像301、302、303和304输入预先建立的安全帽佩戴检测模型，由安全帽佩戴检测模型对待检测压缩图像和待检测压缩子图像进行安全帽佩戴检测，并输出携带安全帽佩戴标记信息的标记图像310和携带安全帽佩戴标记信息的标记子图像306、307、308和309；然后，分别将标记图像310和标记子图像306、307、308和309携带的安全帽佩戴标记信息映射到待检测图像31，可以得到待检测图像31对应的携带安全帽佩戴标记信息的第一目标图像32。For example, as shown in FIG. 3 , after the image to be detected 31 is acquired, the image to be detected can be segmented to obtain multiple sub-images to be detected, and the image to be detected and the multiple sub-images to be detected can be compressed respectively, The
在本申请的一些实施方式中，可以利用安全帽佩戴标记信息在标记图像和标记子图像的图像坐标，将标记图像和标记子图像携带的安全帽佩戴标记信息映射到待检测图像中，得到待检测图像对应的携带安全帽佩戴标记信息的第一目标图像。In some embodiments of the present application, the helmet wearing marking information carried by the marking image and the marking sub-image can be mapped to the image to be detected by using the safety helmet wearing marking information in the image coordinates of the marking image and marking sub-image to obtain the image to be detected. A first target image corresponding to the detection image carrying the marking information for wearing a helmet is detected.
由于标记图像携带的安全帽佩戴标记信息可能与标记子图像携带的安全帽佩戴标记信息重复，因此，在本申请的一些实施方式中，在得到待检测图像对应的携带安全帽佩戴标记信息的第一目标图像之后，可以包括：对第一目标图像中携带的安全帽佩戴标记信息进行过滤，得到去除冗余标记信息的第二目标图像。Since the safety helmet wearing marking information carried by the marked image may overlap with the safety helmet wearing marking information carried by the marking sub-image, in some embodiments of the present application, the first step of obtaining the safety helmet wearing marking information corresponding to the image to be detected is obtained. After a target image, the method may include: filtering the helmet wearing marking information carried in the first target image to obtain a second target image with redundant marking information removed.
例如，如图3所示，可以对第一目标图像32中携带的安全帽佩戴标记信息进行过滤，得到去除冗余标记信息的第二目标图像33。For example, as shown in FIG. 3 , the helmet wearing marking information carried in the
具体的，在本申请的一些实施方式中，可以利用非极大值抑制(Non-MaximumSuppression，NMS)算法，对第一目标图像中携带的安全帽佩戴标记信息进行过滤，得到去除冗余标记信息的第二目标图像。Specifically, in some embodiments of the present application, a Non-Maximum Suppression (NMS) algorithm may be used to filter the helmet wearing marking information carried in the first target image to obtain the removal of redundant marking information the second target image.
本申请实施例中，通过对待检测图像进行分割处理，得到多张待检测子图像，并将待检测图像和待检测子图像输入预先建立的安全帽佩戴检测模型中，不仅通过安全帽佩戴检测模型对待检测图像进行安全帽佩戴检测，使得终端能够检测到在待检测图像中面积占比较大的目标是否佩戴了安全帽，而且由于在待检测图像中面积占比较小的目标在待检测子图像中的面积占比更大，使得终端也能够通过安全帽佩戴检测模型对待检测子图像进行的安全帽佩戴检测，检测到在待检测图像中面积占比较小的目标是否佩戴了安全帽，因此，在分别将安全帽佩戴检测模型输出的标记图像和安全帽佩戴检测模型输出的标记子图像携带的安全帽佩戴标记信息映射到待检测图像后得到的第一目标图像中，不仅包含在待检测图像中面积占比较大的目标对应的安全帽佩戴标记信息，而且包含在待检测图像中面积占比较小的目标对应的安全帽佩戴标记信息，提高了安全帽佩戴检测的精度。In the embodiment of the present application, by dividing the image to be detected, a plurality of sub-images to be detected are obtained, and the images to be detected and the sub-images to be detected are input into the pre-established safety helmet wearing detection model, not only through the safety helmet wearing detection model The helmet wearing detection is performed on the image to be detected, so that the terminal can detect whether the target with a larger area in the image to be detected is wearing a helmet, and because the target with a smaller area in the image to be detected is in the sub-image to be detected. The area of the image to be detected is larger, so that the terminal can also detect the helmet wearing of the sub-image to be detected through the helmet wearing detection model, and detect whether the target with a smaller area in the image to be detected is wearing a helmet. The marking image output by the helmet wearing detection model and the helmet wearing marking information carried by the marking sub-image output by the helmet wearing detection model are respectively mapped to the first target image obtained after the image to be detected, not only included in the image to be detected The helmet wearing mark information corresponding to the target with a larger area ratio is included, and the helmet wearing mark information corresponding to the target with a small area ratio in the image to be detected is included, which improves the accuracy of the helmet wearing detection.
为了方便对上述第二目标图像进行进一步的预览、分析或处理，在本申请的一些实施方式中，在得到去除冗余标记信息的第二目标图像之后，可以包括：判断第二目标图像携带的安全帽佩戴标记信息中是否存在未佩戴安全帽标记信息；若第二目标图像携带的安全帽佩戴标记信息中存在未佩戴安全帽标记信息，则保存第二目标图像。In order to facilitate further previewing, analysis or processing of the second target image, in some embodiments of the present application, after obtaining the second target image with redundant label information removed, the method may include: judging whether the second target image carries the Whether there is no helmet wearing mark information in the helmet wearing mark information; if there is no helmet wearing mark information in the helmet wearing mark information carried by the second target image, save the second target image.
进一步地，可以将上述已保存的第二目标图像的存储路径保存到json文件中，使得前端网页可以根据json文件中保存的存储路径，获取已保存的第二目标图像。Further, the storage path of the above-mentioned saved second target image can be saved in a json file, so that the front-end web page can obtain the saved second target image according to the storage path saved in the json file.
本申请的实施例，通过对第二目标图像携带的安全帽佩戴标记信息中存在未佩戴安全帽标记信息的第二目标图像进行保存，使得安全管理人员可以定期对已保存的第二目标图像进行查看，进而对未佩戴安全帽的人员及时进行安全教育，间接降低安全事故的发生率。In the embodiment of the present application, by saving the second target image that contains the marking information for wearing the helmet without the helmet wearing marking information carried by the second target image, so that the security management personnel can periodically perform the second target image that has been saved. Check, and then carry out safety education for those who do not wear helmets, and indirectly reduce the incidence of safety accidents.
在本申请的另外一些实施方式中，还可以将第二目标图像编码为安全帽佩戴检测视频。In other embodiments of the present application, the second target image may also be encoded as a helmet wearing detection video.
例如，可以利用实时消息传输协议(Real Time Messaging Protocol，rtmp)推流技术，将第二目标图像编码为安全帽佩戴检测视频，并生成该安全帽佩戴检测视频对应的rtmp地址，使得前端网页可以根据该rtmp地址获取到安全帽佩戴检测视频，实现对安全帽佩戴检测视频的预览与播放。For example, the Real Time Messaging Protocol (rtmp) streaming technology can be used to encode the second target image into a helmet wearing detection video, and an rtmp address corresponding to the helmet wearing detection video can be generated, so that the front-end web page can be The helmet wearing detection video is obtained according to the rtmp address, and the preview and playback of the helmet wearing detection video are realized.
需要说明的是，本申请提供的安全帽佩戴检测方法利用预先建立的安全帽佩戴检测模型进行安全帽佩戴检测，检测速度快，因此，在实际应用中，利用监控视频中的每一帧监控图像进行安全帽佩戴检测，并实时地将第二目标图像编码为安全帽佩戴检测视频进行显示，可以实现对监控视频中目标的安全帽佩戴检测情况进行实时显示。It should be noted that the safety helmet wearing detection method provided by the present application uses a pre-established safety helmet wearing detection model to perform safety helmet wearing detection, and the detection speed is fast. Therefore, in practical applications, each frame of monitoring image in the monitoring video is used. The helmet wearing detection is performed, and the second target image is encoded into the helmet wearing detection video in real time for display, which can realize the real-time display of the helmet wearing detection status of the target in the surveillance video.
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本申请并不受所描述的动作顺序的限制，因为根据本申请，某些步骤可以采用其它顺序进行。It should be noted that, for the sake of simple description, the foregoing method embodiments are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the described action sequence. Because according to the present application, certain steps may be performed in other orders.
如图4所示为本申请实施例提供一种安全帽佩戴检测装置400的结构示意图，所述安全帽佩戴检测装置400可以包括：获取单元401、分割单元402、检测单元403和映射单元404。FIG. 4 is a schematic structural diagram of a safety helmet wearing
获取单元401，用于获取待检测图像；an
分割单元402，用于对所述待检测图像进行分割处理，得到多张待检测子图像；A
检测单元403，用于将所述待检测图像和所述待检测子图像输入预先建立的安全帽佩戴检测模型，由所述安全帽佩戴检测模型对所述待检测图像和所述待检测子图像进行安全帽佩戴检测，并输出携带安全帽佩戴标记信息的标记图像和携带安全帽佩戴标记信息的标记子图像；The
映射单元404，用于分别将所述标记图像和所述标记子图像携带的安全帽佩戴标记信息映射到所述待检测图像，得到所述待检测图像对应的携带安全帽佩戴标记信息的第一目标图像。The
在本申请的一些实施方式中，上述映射单元404还具体用于：对所述第一目标图像中携带的安全帽佩戴标记信息进行过滤，得到去除冗余标记信息的第二目标图像。In some embodiments of the present application, the
在本申请的一些实施方式中，上述安全帽佩戴检测装置还包括存储单元，用于：判断所述第二目标图像携带的安全帽佩戴标记信息中是否存在未佩戴安全帽标记信息；若所述第二目标图像携带的安全帽佩戴标记信息中存在未佩戴安全帽标记信息，则保存所述第二目标图像。In some embodiments of the present application, the above-mentioned safety helmet wearing detection device further includes a storage unit for: judging whether there is no safety helmet marking information in the safety helmet wearing marking information carried by the second target image; If there is no helmet wearing mark information in the helmet wearing mark information carried by the second target image, the second target image is saved.
在本申请的一些实施方式中，上述存储单元还具体用于：将所述第二目标图像编码为安全帽佩戴检测视频。In some embodiments of the present application, the above-mentioned storage unit is further specifically configured to: encode the second target image into a helmet wearing detection video.
在本申请的一些实施方式中，上述检测单元403还具体用于：对所述待检测图像和所述待检测子图像分别进行压缩处理，得到预设分辨率的待检测压缩图像和待检测压缩子图像，并将所述待检测压缩图像和所述待检测压缩子图像输入预先建立的安全帽佩戴检测模型。In some embodiments of the present application, the above-mentioned
在本申请的一些实施方式中，上述检测单元403还具体用于：对待训练的安全帽佩戴检测模型进行训练，得到预先建立的安全帽佩戴检测模型；其中，所述待训练的安全帽佩戴检测模型的训练，包括：获取多个样本图像，以及分别与每个样本图像对应的携带预先标记的安全帽佩戴标记信息的标准图像；将所述多个样本图像中的目标样本图像输入所述待训练的安全帽佩戴检测模型中，由所述待训练的安全帽佩戴检测模型输出所述目标样本图像对应的携带安全帽佩戴标记信息的待确认图像；计算所述待确认图像中的安全帽佩戴标记信息与所述目标样本图像对应的标准图像携带的预先标记的安全帽佩戴标记信息之间的匹配度，若所述匹配度小于匹配度阈值，则调整所述待训练的安全帽佩戴检测模型的参数，重新利用所述目标样本图像对所述待训练的安全帽佩戴检测模型进行训练，直至利用所述目标样本图像对所述待训练的安全帽佩戴检测模型进行训练的次数大于或等于第一次数阈值时，或者，所述待确认图像中的安全帽佩戴标记信息与所述目标样本图像对应的标准图像携带的预先标记的安全帽佩戴标记信息之间的匹配度大于或等于匹配度阈值时，利用所述多个样本图像中的下一个目标样本图像对所述待训练的安全帽佩戴检测模型进行训练，直至所述待训练的安全帽佩戴检测模型的总训练次数大于或等于第二次数阈值，得到所述预先建立的安全帽佩戴检测模型。In some embodiments of the present application, the above-mentioned
在本申请的一些实施方式中，上述检测单元403还具体用于：对所述预先建立的安全帽佩戴检测模型中的参数进行量化处理，得到量化处理后的安全帽佩戴检测模型。In some embodiments of the present application, the
需要说明的是，为描述的方便和简洁，上述安全帽佩戴检测装置400的具体工作过程，可以参考图1至图3所述方法的对应过程，在此不再赘述。It should be noted that, for the convenience and brevity of description, the specific working process of the above-mentioned safety helmet wearing
如图5所示，为本申请实施例提供的一种终端的示意图。该终端5可以包括：处理器50、存储器51以及存储在所述存储器51中并可在所述处理器50上运行的计算机程序52，例如安全帽佩戴检测装置程序。所述处理器50执行所述计算机程序52时实现上述各个安全帽佩戴检测方法实施例中的步骤，例如图1所示的步骤101至104。或者，所述处理器50执行所述计算机程序52时实现上述各装置实施例中各模块/单元的功能，例如图4所示单元401至404的功能。As shown in FIG. 5 , it is a schematic diagram of a terminal according to an embodiment of the present application. The
所述计算机程序可以被分割成一个或多个模块/单元，所述一个或者多个模块/单元被存储在所述存储器51中，并由所述处理器50执行，以完成本申请。所述一个或多个模块/单元可以是能够完成特定功能的一系列计算机程序指令段，该指令段用于描述所述计算机程序在所述终端中的执行过程。例如，所述计算机程序可以被分割成获取单元、分割单元、检测单元和映射单元，各单元具体功能如下：获取单元，用于获取待检测图像；分割单元，用于对所述待检测图像进行分割处理，得到多张待检测子图像；检测单元，用于将所述待检测图像和所述待检测子图像输入预先建立的安全帽佩戴检测模型，由所述安全帽佩戴检测模型对所述待检测图像和所述待检测子图像进行安全帽佩戴检测，并输出携带安全帽佩戴标记信息的标记图像和携带安全帽佩戴标记信息的标记子图像；映射单元，用于分别将所述标记图像和所述标记子图像携带的安全帽佩戴标记信息映射到所述待检测图像，得到所述待检测图像对应的携带安全帽佩戴标记信息的第一目标图像。The computer program may be divided into one or more modules/units, which are stored in the
所述终端可以是智能电视等移动终端，或者是智能手机、桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述终端可包括，但不仅限于，处理器50、存储器51。本领域技术人员可以理解，图5仅仅是终端的示例，并不构成对终端的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如所述终端还可以包括输入输出设备、网络接入设备、总线等。The terminal may be a mobile terminal such as a smart TV, or a computing device such as a smart phone, a desktop computer, a notebook, a palmtop computer, and a cloud server. The terminal may include, but is not limited to, the
所称处理器50可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字安全帽佩戴检测器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。The so-called
所述存储器51可以是所述终端的内部存储单元，例如终端的硬盘或内存。所述存储器51也可以是所述终端的外部存储设备，例如所述终端上配备的插接式硬盘，智能存储卡(Smart Media Card，SMC)，安全数字(Secure Digital，SD)卡，闪存卡(Flash Card)等。进一步地，所述存储器51还可以既包括所述终端的内部存储单元也包括外部存储设备。所述存储器51用于存储所述计算机程序以及所述终端所需的其他程序和数据。所述存储器51还可以用于暂时地存储已经输出或者将要输出的数据。The
所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，仅以上述各功能单元、模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能单元、模块完成，即将所述装置的内部结构划分成不同的功能单元或模块，以完成以上描述的全部或者部分功能。实施例中的各功能单元、模块可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中，上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。另外，各功能单元、模块的具体名称也只是为了便于相互区分，并不用于限制本申请的保护范围。上述系统中单元、模块的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. Module completion, that is, dividing the internal structure of the device into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment may be integrated in one processing unit, or each unit may exist physically alone, or two or more units may be integrated in one unit, and the above-mentioned integrated units may adopt hardware. It can also be realized in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing from each other, and are not used to limit the protection scope of the present application. For the specific working processes of the units and modules in the above-mentioned system, reference may be made to the corresponding processes in the foregoing method embodiments, which will not be repeated here.
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述或记载的部分，可以参见其它实施例的相关描述。In the foregoing embodiments, the description of each embodiment has its own emphasis. For parts that are not described or described in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of this application.
在本申请所提供的实施例中，应该理解到，所揭露的装置/终端和方法，可以通过其它的方式实现。例如，以上所描述的装置/终端实施例仅仅是示意性的，例如，所述模块或单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通讯连接可以是通过一些接口，装置或单元的间接耦合或通讯连接，可以是电性，机械或其它的形式。In the embodiments provided in this application, it should be understood that the disclosed apparatus/terminal and method may be implemented in other manners. For example, the device/terminal embodiments described above are only illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or Components may be combined or may be integrated into another system, or some features may be omitted, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。The units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.
所述集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请实现上述实施例方法中的全部或部分流程，也可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一计算机可读存储介质中，该计算机程序在被处理器执行时，可实现上述各个方法实施例的步骤。其中，所述计算机程序包括计算机程序代码，所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质可以包括：能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(Read-Only Memory，ROM)、随机存取存储器(RandomAccess Memory，RAM)、电载波信号、电信信号以及软件分发介质等。需要说明的是，所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减，例如在某些司法管辖区，根据立法和专利实践，计算机可读介质不包括电载波信号和电信信号。The integrated modules/units, if implemented in the form of software functional units and sold or used as independent products, may be stored in a computer-readable storage medium. Based on this understanding, the present application can implement all or part of the processes in the methods of the above embodiments, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer When the program is executed by the processor, the steps of the foregoing method embodiments can be implemented. Wherein, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form, and the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, removable hard disk, magnetic disk, optical disk, computer memory, Read-Only Memory (ROM) , Random Access Memory (Random Access Memory, RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable media may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer-readable media Electric carrier signals and telecommunication signals are not included.
以上所述实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围，均应包含在本申请的保护范围之内。The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the above-mentioned embodiments, those of ordinary skill in the art should understand that: it can still be used for the above-mentioned implementations. The technical solutions described in the examples are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions in the embodiments of the application, and should be included in the within the scope of protection of this application.
Claims (10)
Translated fromChinesePriority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010220020.5ACN111476117A (en) | 2020-03-25 | 2020-03-25 | Safety helmet wearing detection method and device and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010220020.5ACN111476117A (en) | 2020-03-25 | 2020-03-25 | Safety helmet wearing detection method and device and terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111476117Atrue CN111476117A (en) | 2020-07-31 |
Family
ID=71748403
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010220020.5APendingCN111476117A (en) | 2020-03-25 | 2020-03-25 | Safety helmet wearing detection method and device and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111476117A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112487976A (en)* | 2020-11-30 | 2021-03-12 | 中科院计算所西部高等技术研究院 | Monitoring method and device based on image recognition and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222672A (en)* | 2019-06-19 | 2019-09-10 | 广东工业大学 | The safety cap of construction site wears detection method, device, equipment and storage medium |
| CN110310264A (en)* | 2019-06-25 | 2019-10-08 | 北京邮电大学 | A large-scale target detection method and device based on DCNN |
| CN110443765A (en)* | 2019-08-02 | 2019-11-12 | 厦门美图之家科技有限公司 | Image processing method, device and electronic equipment |
| CN110766650A (en)* | 2019-08-05 | 2020-02-07 | 南方科技大学 | Biological detection early warning method, system, device, computer equipment and storage medium |
| CN110889376A (en)* | 2019-11-28 | 2020-03-17 | 创新奇智(南京)科技有限公司 | Safety helmet wearing detection system and method based on deep learning |
- 2020
- 2020-03-25CNCN202010220020.5Apatent/CN111476117A/enactivePending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222672A (en)* | 2019-06-19 | 2019-09-10 | 广东工业大学 | The safety cap of construction site wears detection method, device, equipment and storage medium |
| CN110310264A (en)* | 2019-06-25 | 2019-10-08 | 北京邮电大学 | A large-scale target detection method and device based on DCNN |
| CN110443765A (en)* | 2019-08-02 | 2019-11-12 | 厦门美图之家科技有限公司 | Image processing method, device and electronic equipment |
| CN110766650A (en)* | 2019-08-05 | 2020-02-07 | 南方科技大学 | Biological detection early warning method, system, device, computer equipment and storage medium |
| CN110889376A (en)* | 2019-11-28 | 2020-03-17 | 创新奇智(南京)科技有限公司 | Safety helmet wearing detection system and method based on deep learning |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112487976A (en)* | 2020-11-30 | 2021-03-12 | 中科院计算所西部高等技术研究院 | Monitoring method and device based on image recognition and storage medium |
| CN112487976B (en)* | 2020-11-30 | 2023-10-24 | 中科院计算所西部高等技术研究院 | Monitoring method, device and storage medium based on image recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110147717B (en) | Human body action recognition method and device | |
| CN109670441B (en) | Method, system, terminal and computer readable storage medium for realizing wearing recognition of safety helmet | |
| CN109657564B (en) | Personnel on-the-job detection method, device, storage medium and terminal device | |
| CN110163143A (en) | Unlawful practice recognition methods, device and terminal device | |
| CN114937232B (en) | Method, system and equipment for detecting the wearing of protective equipment for medical waste disposal personnel | |
| CN109948420B (en) | Face comparison method and device and terminal equipment | |
| CN110096945A (en) | Indoor Video key frame of video real time extracting method based on machine learning | |
| CN108174198B (en) | Video image quality diagnosis analysis detection device and application system | |
| CN114926778A (en) | Safety helmet and personnel identity recognition system under production environment | |
| CN115131714A (en) | Video image intelligent detection and analysis method and system | |
| CN113869137B (en) | Event detection method, device, terminal equipment and storage medium | |
| CN108304831A (en) | A kind of method and device that monitoring worker safety helmet is worn | |
| CN109993021A (en) | Face detection method, device and electronic device | |
| CN111967319A (en) | Infrared and visible light based in-vivo detection method, device, equipment and storage medium | |
| CN108288025A (en) | A kind of car video monitoring method, device and equipment | |
| CN111223011A (en) | A method and system for food safety supervision of catering enterprises based on video analysis | |
| CN117854402A (en) | Abnormal display detection method and device of display screen and terminal equipment | |
| WO2022193516A1 (en) | Depth camera-based pedestrian flow analysis method and apparatus | |
| CN118351487A (en) | Employee violation behavior identification method based on AI | |
| CN111597985A (en) | Dynamic identification method and device for equipment wearing and electronic equipment | |
| CN111476117A (en) | Safety helmet wearing detection method and device and terminal | |
| CN113255472B (en) | Face quality evaluation method and system based on random embedding stability | |
| CN115995093A (en) | Safety helmet wearing identification method based on improved YOLOv5 | |
| CN112183359A (en) | Violent content detection method, device and device in video | |
| WO2025020442A1 (en) | Video feature extraction method and apparatus, readable storage medium, and terminal device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20200731 | |
| RJ01 | Rejection of invention patent application after publication |