CN114170185A - Camera detection method and system based on visual information - Google Patents
Camera detection method and system based on visual informationDownload PDFInfo
- Publication number
- CN114170185A CN114170185ACN202111492771.3ACN202111492771ACN114170185ACN 114170185 ACN114170185 ACN 114170185ACN 202111492771 ACN202111492771 ACN 202111492771ACN 114170185 ACN114170185 ACN 114170185A
- Authority
- CN
- China
- Prior art keywords
- camera
- image
- detection result
- detection
- information
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
Description
Translated fromChinese技术领域technical field
本发明涉及相机检测技术领域，具体涉及一种基于视觉信息的相机检测方法及系统。The invention relates to the technical field of camera detection, in particular to a camera detection method and system based on visual information.
背景技术Background technique
随着移动机器人的发展，基于视觉传感器的移动机器逐渐成为机器人领域的一大发展方向，然而基于视觉算法的移动机器人容易受到视觉传感器采集到图像的质量及周围环境的影响，如果相机被遮挡或受环境影响导致图像质量过低，会导致视觉算法失效或得到错误结果。With the development of mobile robots, mobile machines based on vision sensors have gradually become a major development direction in the field of robotics. However, mobile robots based on vision algorithms are easily affected by the quality of images collected by vision sensors and the surrounding environment. If the camera is blocked or Affected by the environment, the image quality is too low, which can cause the vision algorithm to fail or get erroneous results.
发明内容SUMMARY OF THE INVENTION
本发明的目的在于提供一种基于视觉信息的相机检测方法及系统。以期解决背景技术中存在的技术问题。The purpose of the present invention is to provide a camera detection method and system based on visual information. In order to solve the technical problems existing in the background technology.
为了实现上述目的，本发明采用以下技术方案：In order to achieve the above object, the present invention adopts the following technical solutions:
一种基于视觉信息的相机检测方法，包括：A camera detection method based on visual information, comprising:
获取由待检测相机采集的环境图像序列；Obtain the sequence of environmental images collected by the camera to be detected;
基于对所述环境图像序列获取候选图像及其图像信息；Obtaining candidate images and their image information based on the sequence of environmental images;
基于所述图像信息确定所述待检测相机的第一检测结果；所述第一检测结果与光照环境相关；Determine the first detection result of the camera to be detected based on the image information; the first detection result is related to the lighting environment;
基于所述图像信息确定所述待检测相机的第二检测结果；所述第二检测结果与遮挡情况相关；A second detection result of the camera to be detected is determined based on the image information; the second detection result is related to an occlusion situation;
对所述候选图像进行预处理，基于检测模型对预处理后的候选图像的处理，得到所述检测相机的第三检测结果；所述第三检测结果与遮挡情况相关；Preprocessing the candidate image, and processing the preprocessed candidate image based on the detection model to obtain a third detection result of the detection camera; the third detection result is related to the occlusion situation;
结合所述第一检测结果、所述第二检测结果确定所述第三检测结果的误判率，并得到所述待检测相机的最终检测结果。The misjudgment rate of the third detection result is determined by combining the first detection result and the second detection result, and the final detection result of the camera to be detected is obtained.
在一些实施例中，所述图像信息至少包括以下一种：所述候选图像成像时的曝光程度和进光量、所述候选图像的像素灰度和梯度的位置分布差异。In some embodiments, the image information includes at least one of the following: exposure degree and light input amount of the candidate image during imaging, and the positional distribution difference of pixel grayscale and gradient of the candidate image.
在一些实施例中，所述基于所述图像信息确定所述待检测相机的第一检测结果包括：In some embodiments, the determining the first detection result of the camera to be detected based on the image information includes:
基于所述候选图像成像时的曝光程度和进光量，判断所述待检测相机是否处于强光或弱光环境。Based on the exposure degree and the amount of incoming light when the candidate image is formed, it is determined whether the camera to be detected is in a strong light or weak light environment.
在一些实施例中，所述基于所述图像信息确定所述待检测相机的第二检测结果包括：In some embodiments, the determining the second detection result of the camera to be detected based on the image information includes:
基于所述候选图像的像素灰度和梯度的位置分布差异，计算潜在遮挡的几何分布，根据分布差异的统计值，统计潜在遮挡的聚合程度。The geometric distribution of potential occlusions is calculated based on the positional distribution differences of pixel grayscales and gradients of the candidate images, and the aggregation degree of potential occlusions is calculated according to the statistical value of the distribution differences.
在一些实施例中，所述对所述候选图像进行预处理，基于检测模型对预处理后的候选图像的处理，得到所述检测相机的第三检测结果包括：In some embodiments, the preprocessing of the candidate image and the processing of the preprocessed candidate image based on the detection model to obtain the third detection result of the detection camera include:
对所述候选图像进行信息缩减，并假定所述候选图像中存在遮挡，使用非监督的机器学习自推演算法计算该遮挡的位置信息和灰度信息，判断该假定遮挡是否存在。Perform information reduction on the candidate image, and assume that there is occlusion in the candidate image, use an unsupervised machine learning self-derivation algorithm to calculate the position information and grayscale information of the occlusion, and determine whether the assumed occlusion exists.
在一些实施例中，所述判断该假定遮挡是否存在包括：利用遮挡前景与背景所在像素的位置和灰度差异性，判断是否存在遮挡物体。In some embodiments, the judging whether the assumed occlusion exists includes: judging whether there is an occlusion object by using the position and grayscale difference of the pixels where the occlusion foreground and the background are located.
在一些实施例中，所述结合所述第一检测结果、所述第二检测结果确定所述第三检测结果的误判率，并得到所述待检测相机的最终检测结果包括：In some embodiments, determining the false positive rate of the third detection result by combining the first detection result and the second detection result, and obtaining the final detection result of the camera to be detected includes:
根据环境图像序列中的弱光和遮挡结果变化序列，计算当前图像检测判定的误判率，并给出最终的所述待检测相机的遮挡和强弱光环境检测结果。According to the change sequence of weak light and occlusion results in the environmental image sequence, the misjudgment rate of the current image detection determination is calculated, and the final occlusion and strong and weak light environment detection results of the camera to be detected are given.
在一些实施例中，所述根据环境图像序列中的弱光和遮挡结果变化序列，计算当前图像检测判定的误判率，并给出最终的所述待检测相机的遮挡和强弱光环境检测结果包括：对连续多帧环境图像序列进行相应处理获得所述第一检测结果、所述第二检测结果、所述第三检测结果；根据对连续多帧环境图像序列的判断结果的统计概率得出当前图像序列最终的相机遮挡和弱光环境检测结果。In some embodiments, according to the sequence of changes in the weak light and occlusion results in the environmental image sequence, the misjudgment rate of the current image detection determination is calculated, and the final occlusion and strong and weak light environment detection of the camera to be detected are given. The results include: correspondingly processing the continuous multi-frame environmental image sequence to obtain the first detection result, the second detection result, and the third detection result; according to the statistical probability of the judgment results of the continuous multi-frame environmental image sequence The final camera occlusion and low-light environment detection results of the current image sequence are obtained.
同时，本发明还公开了一种基于视觉信息的相机检测系统，包括：Meanwhile, the present invention also discloses a camera detection system based on visual information, comprising:
图像获取模块，用于获取由待检测相机采集的环境图像序列及基于对所述环境图像序列获取候选图像；an image acquisition module, configured to acquire a sequence of environmental images collected by the camera to be detected and to acquire candidate images based on the sequence of environmental images;
成像信息统计模块，用于统计计算候选图像整体亮度和清晰度，分析候选图像成像时的曝光程度和进光量，并基于所述成像时的曝光程度和进光量，判断待检测相机是否处于强光或弱光环境中；The imaging information statistics module is used to statistically calculate the overall brightness and clarity of the candidate image, analyze the exposure degree and light input amount of the candidate image during imaging, and determine whether the camera to be detected is in strong light based on the exposure degree and light input amount during imaging. or in a low-light environment;
潜在遮挡分布计算模块，用于计算候选图像像素灰度和梯度的位置分布差异，获得潜在遮挡的几何分布，根据分布差异的统计值，统计潜在遮挡的聚合程度；The potential occlusion distribution calculation module is used to calculate the position distribution difference of the pixel grayscale and gradient of the candidate image, obtain the geometric distribution of the potential occlusion, and calculate the aggregation degree of the potential occlusion according to the statistical value of the distribution difference;
遮挡真实性判断模块，用于对候选图像进行信息缩减，并假定图像中存在遮挡，使用非监督的机器学习自推演算法计算该遮挡的位置信息和灰度信息，判断该假定遮挡是否存在；The occlusion authenticity judgment module is used to reduce the information of the candidate image, and it is assumed that there is occlusion in the image, and the unsupervised machine learning self-derivation algorithm is used to calculate the position information and gray level information of the occlusion, and determine whether the assumed occlusion exists;
误判率检查模块，用于根据环境图像序列中的强弱光和遮挡结果变化序列，计算当前图像检测判定的误判率，并给出最终的相机遮挡和强光或弱光环境检测结果。The false positive rate checking module is used to calculate the false positive rate of the current image detection and judgment according to the strong and weak light and occlusion result change sequence in the environmental image sequence, and give the final camera occlusion and strong light or weak light environment detection results.
同时，本发明还公开了一种基于视觉信息的相机检测装置，所述装置包括处理器以及存储器；所述存储器用于存储指令，所述指令被所述处理器执行时，导致所述装置实现上述任一项所述基于视觉信息的相机检测方法。At the same time, the present invention also discloses a camera detection device based on visual information, the device includes a processor and a memory; the memory is used for storing instructions, and when the instructions are executed by the processor, the device is implemented The camera detection method based on visual information according to any one of the above.
同时，本发明还公开了一种计算机可读存储介质，所述存储介质存储计算机指令，当计算机读取存储介质中的计算机指令后，计算机运行上述任一项所述基于视觉信息的相机检测方法。At the same time, the present invention also discloses a computer-readable storage medium, the storage medium stores computer instructions, and after the computer reads the computer instructions in the storage medium, the computer executes any of the above-mentioned visual information-based camera detection methods .
有益效果beneficial effect
本发明与现有技术相比，其显著优点是：Compared with the prior art, the present invention has the following significant advantages:
通过计算图像亮度及清晰度、潜在遮挡的几何分布及机器学习聚类多层筛选，对相机遮挡和强光或弱光环境进行检测并进行真实性和误判检查，可以解决视觉移动机器人算法容易受到视觉传感器采集到图像的质量及周围环境的影响导致算法失效或结果错误的问题，提高视觉移动机器人算法的精确度和鲁棒性。By calculating the image brightness and clarity, the geometric distribution of potential occlusion, and the multi-layer screening of machine learning clustering, the camera occlusion and the strong or weak light environment are detected, and the authenticity and misjudgment check can be solved. Influenced by the quality of the image collected by the vision sensor and the surrounding environment, the algorithm fails or the result is wrong, and the accuracy and robustness of the visual mobile robot algorithm are improved.
附图说明Description of drawings
图1是本实施例涉及基于视觉信息的相机检测系统示意图；FIG. 1 is a schematic diagram of a camera detection system based on visual information according to this embodiment;
图2是本实施例涉及的基于视觉信息的相机检测方法流程示意图。FIG. 2 is a schematic flowchart of the camera detection method based on visual information involved in this embodiment.
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。In order to make the objectives, technical solutions and advantages of the present application more clear, the present application will be further described in 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.
相反，本申请涵盖任何由权利要求定义的在本申请的精髓和范围上做的替代、修改、等效方法以及方案。进一步，为了使公众对本申请有更好的了解，在下文对本申请的细节描述中，详尽描述了一些特定的细节部分。对本领域技术人员来说没有这些细节部分的描述也可以完全理解本申请。On the contrary, this application covers any alternatives, modifications, equivalents and arrangements within the spirit and scope of this application as defined by the claims. Further, in order for the public to have a better understanding of the present application, some specific details are described in detail in the following detailed description of the present application. Those skilled in the art can fully understand the present application without the description of these detailed parts.
以下将结合图1-2对本申请实施例所涉及的一种基于视觉信息的相机检测方法进行详细说明。值得注意的是，以下实施例仅仅用于解释本申请，并不构成对本申请的限定。A visual information-based camera detection method involved in the embodiments of the present application will be described in detail below with reference to FIGS. 1-2 . It should be noted that the following examples are only used to explain the present application, and do not constitute a limitation to the present application.
实施例1Example 1
如图1所示，一种基于视觉信息的相机检测系统100，包括：As shown in FIG. 1, a
图像获取模块101，用于获取由待检测相机采集的环境图像序列及基于对所述环境图像序列获取候选图像；An image acquisition module 101, configured to acquire a sequence of environmental images collected by a camera to be detected and to acquire candidate images based on the sequence of environmental images;
成像信息统计模块102，用于统计计算候选图像整体亮度和清晰度，分析候选图像成像时的曝光程度和进光量，并基于所述成像时的曝光程度和进光量，判断待检测相机是否处于强光或弱光环境中；The imaging information statistics module 102 is used to statistically calculate the overall brightness and clarity of the candidate image, analyze the exposure degree and light input amount of the candidate image when imaging, and determine whether the camera to be detected is in a strong state based on the exposure degree and light input amount during imaging. light or low light environment;
潜在遮挡分布计算模块103，用于计算候选图像像素灰度和梯度的位置分布差异，获得潜在遮挡的几何分布，根据分布差异的统计值，统计潜在遮挡的聚合程度；The potential occlusion distribution calculation module 103 is used to calculate the position distribution difference of the pixel grayscale and gradient of the candidate image, obtain the geometric distribution of the potential occlusion, and count the aggregation degree of the potential occlusion according to the statistical value of the distribution difference;
遮挡真实性判断模块104，用于对候选图像进行信息缩减，并假定图像中存在遮挡，使用非监督的机器学习自推演算法计算该遮挡的位置信息和灰度信息，判断该假定遮挡是否存在；The occlusion authenticity judgment module 104 is used to reduce the information of the candidate image, and assuming that there is occlusion in the image, use an unsupervised machine learning self-derivation algorithm to calculate the position information and gray level information of the occlusion, and determine whether the assumed occlusion exists;
误判率检查模块105，用于根据环境图像序列中的强弱光和遮挡结果变化序列，计算当前图像检测判定的误判率，并给出最终的相机遮挡和强光或弱光环境检测结果。The false positive rate checking module 105 is used to calculate the false positive rate of the current image detection and judgment according to the strong and weak light and the occlusion result change sequence in the environmental image sequence, and give the final camera occlusion and strong light or weak light environment detection results .
如图2所示为一种基于视觉信息的相机检测方法包括的流程，流程200包括：As shown in FIG. 2 , a process included in a camera detection method based on visual information, the
步骤210，获取由待检测相机采集的环境图像序列。Step 210: Obtain a sequence of environmental images collected by the camera to be detected.
在一些实施例中，可以基于视觉传感器获取环境图像序列。In some embodiments, a sequence of environmental images may be acquired based on a vision sensor.
步骤220，基于对所述环境图像序列获取候选图像及其图像信息；在一些实施例中，所述图像信息至少包括以下一种：所述候选图像成像时的曝光程度和进光量、所述候选图像的像素灰度和梯度的位置分布差异。Step 220: Obtain a candidate image and its image information based on the sequence of environmental images; in some embodiments, the image information includes at least one of the following: the exposure degree and the amount of incoming light when the candidate image is imaged, the candidate image Differences in the positional distribution of pixel grayscales and gradients of an image.
在一些实施例中，可以通过对所述环境图像序列进行图像信息提取，计算图像整体亮度和清晰度，从而计算成像时的曝光程度和进光量。In some embodiments, the overall brightness and sharpness of the image can be calculated by extracting image information from the environmental image sequence, so as to calculate the exposure degree and the amount of incoming light during imaging.
步骤230，基于所述图像信息确定所述待检测相机的第一检测结果；所述第一检测结果与光照环境相关。Step 230: Determine a first detection result of the camera to be detected based on the image information; the first detection result is related to the lighting environment.
在一些实施例中，步骤230具体包括基于所述候选图像成像时的曝光程度和进光量，判断所述待检测相机是否处于强光或弱光环境。In some embodiments, step 230 specifically includes judging whether the camera to be detected is in a strong light or weak light environment based on the exposure degree and the amount of incoming light when the candidate image is formed.
在一些实施例中，步骤230具体包括对所述环境图像序列进行图像信息提取，计算图像整体亮度和清晰度，从而计算成像时的曝光程度和进光量，判断是否处于强光或弱光环境中，对候选图像进行第一次筛选。In some embodiments, step 230 specifically includes extracting image information from the environmental image sequence, calculating the overall brightness and sharpness of the image, thereby calculating the exposure degree and the amount of incoming light during imaging, and judging whether it is in a strong light or weak light environment , perform the first screening of candidate images.
在一些实施例中，对所述环境图像进行成像信息分析的方法，包括但不限于统计计算图像整体亮度、梯度阈值筛选、模拟曝光程度信息等方法，判断图像是否处于强光或弱光环境中，或可能存在较大范围相机遮挡，滤除质量较低的图片。In some embodiments, the method for analyzing the imaging information of the environmental image includes, but is not limited to, methods such as statistical calculation of the overall brightness of the image, gradient threshold screening, simulated exposure level information, etc., to determine whether the image is in a strong light or weak light environment. , or there may be a large range of camera occlusions, filtering out lower-quality images.
步骤240，基于所述图像信息确定所述待检测相机的第二检测结果；所述第二检测结果与遮挡情况相关。Step 240: Determine a second detection result of the camera to be detected based on the image information; the second detection result is related to an occlusion situation.
在一些实施例中，步骤240具体包括：基于所述候选图像的像素灰度和梯度的位置分布差异，计算潜在遮挡的几何分布，根据分布差异的统计值，统计潜在遮挡的聚合程度。In some embodiments, step 240 specifically includes: calculating the geometric distribution of potential occlusions based on the positional distribution differences of pixel grayscales and gradients of the candidate images, and calculating the aggregation degree of potential occlusions according to the statistical value of the distribution differences.
在一些实施例中，步骤240可以通过以下方式实现：计算候选图像像素灰度和梯度的位置分布差异，从而计算潜在遮挡的几何分布，根据分布差异的统计值，统计潜在遮挡的聚合程度，进行第二次筛选。其中，In some embodiments,
在一些实施例中，对所述环境图像进行潜在遮挡的聚合程度分析方法包括但不限于灰度直方图、Canny边缘检测算法、Laplacian算子边缘检测算法以及其他通过图像梯度信息获取边缘的算法等。In some embodiments, the method for analyzing the aggregation degree of potential occlusion on the environmental image includes, but is not limited to, grayscale histogram, Canny edge detection algorithm, Laplacian operator edge detection algorithm, and other algorithms that obtain edges through image gradient information, etc. .
步骤250，对所述候选图像进行预处理，基于检测模型对预处理后的候选图像的处理，得到所述检测相机的第三检测结果；所述第三检测结果与遮挡情况相关。
在一些实施例中，所步骤250具体包括：In some embodiments, the
对所述候选图像进行信息缩减，并假定所述候选图像中存在遮挡，使用非监督的机器学习自推演算法计算该遮挡的位置信息和灰度信息，判断该假定遮挡是否存在。其中，对所述环境图像的潜在遮挡的聚合程度分析方法包括但不限于基于灰度的二值化方法、基于梯度的分块方差统计方法等。Perform information reduction on the candidate image, and assume that there is occlusion in the candidate image, use an unsupervised machine learning self-derivation algorithm to calculate the position information and grayscale information of the occlusion, and determine whether the assumed occlusion exists. Wherein, the method for analyzing the aggregation degree of the potential occlusion of the environment image includes, but is not limited to, a grayscale-based binarization method, a gradient-based block variance statistics method, and the like.
在一些实施例中，所述判断该假定遮挡是否存在包括：利用遮挡前景与背景所在像素的位置和灰度差异性，判断是否存在遮挡物体。In some embodiments, the judging whether the assumed occlusion exists includes: judging whether there is an occlusion object by using the position and grayscale difference of the pixels where the occlusion foreground and the background are located.
在一些实施例中，步骤250可以通过以下方式实现：对候选图像进行信息缩减，并假定图像中存在遮挡，使用非监督的机器学习自推演算法计算该遮挡的位置信息和灰度信息，判断该假定遮挡是否存在，进行第三次筛选.In some embodiments, step 250 can be implemented by the following methods: performing information reduction on the candidate image, and assuming that there is occlusion in the image, using an unsupervised machine learning self-derivation algorithm to calculate the position information and grayscale information of the occlusion, and determining the Assuming the existence of occlusion, a third screening is performed.
在一些实施例中，非监督的机器学习自推演算法包括K-均值聚类、凝聚层次聚类、噪声密度聚类等方法，确定机器学习定位后的遮挡的真实性方法包括，利用遮挡前景与背景所在像素的位置和灰度差异性，判断结果确实是否为遮挡物体。In some embodiments, the unsupervised machine learning self-derivation algorithm includes methods such as K-means clustering, agglomerative hierarchical clustering, noise density clustering, etc., and the method for determining the authenticity of the occlusion located by the machine learning includes using the occlusion foreground and The position and grayscale difference of the pixel where the background is located determines whether the result is indeed an occluding object.
步骤260，结合所述第一检测结果、所述第二检测结果确定所述第三检测结果的误判率，并得到所述待检测相机的最终检测结果。
在一些实施例中，步骤260具体包括：In some embodiments, step 260 specifically includes:
根据环境图像序列中的弱光和遮挡结果变化序列，计算当前图像检测判定的误判率，并给出最终的所述待检测相机的遮挡和强弱光环境检测结果。According to the change sequence of weak light and occlusion results in the environmental image sequence, the misjudgment rate of the current image detection determination is calculated, and the final occlusion and strong and weak light environment detection results of the camera to be detected are given.
在一些实施例中，所述根据环境图像序列中的弱光和遮挡结果变化序列，计算当前图像检测判定的误判率，并给出最终的所述待检测相机的遮挡和强弱光环境检测结果包括：对连续多帧环境图像序列进行相应处理获得所述第一检测结果、所述第二检测结果、所述第三检测结果；根据对连续多帧环境图像序列的判断结果的统计概率得出当前图像序列最终的相机遮挡和弱光环境检测结果。In some embodiments, according to the sequence of changes in the weak light and occlusion results in the environmental image sequence, the misjudgment rate of the current image detection determination is calculated, and the final occlusion and strong and weak light environment detection of the camera to be detected are given. The results include: correspondingly processing the continuous multi-frame environmental image sequence to obtain the first detection result, the second detection result, and the third detection result; according to the statistical probability of the judgment results of the continuous multi-frame environmental image sequence The final camera occlusion and low-light environment detection results of the current image sequence are obtained.
在一些实施例中，步骤250可以通过以下方式实现：根据前序图像序列中的弱光和遮挡结果变化序列，计算当前图像检测判定的误判率，并给出最终的相机遮挡和强弱光环境检测结果。即对连续多帧图像序列进行所述三次筛选，根据多帧判断结果的统计概率得出当前图像序列最终的相机遮挡和弱光环境检测结果。In some embodiments,
同时，本发明还公开了一种基于视觉信息的相机检测装置，所述装置包括处理器以及存储器；所述存储器用于存储指令，所述指令被所述处理器执行时，导致所述装置实现上述任一项所述基于视觉信息的相机检测方法。At the same time, the present invention also discloses a camera detection device based on visual information, the device includes a processor and a memory; the memory is used for storing instructions, and when the instructions are executed by the processor, the device is implemented The camera detection method based on visual information according to any one of the above.
同时，本发明还公开了一种计算机可读存储介质，所述存储介质存储计算机指令，当计算机读取存储介质中的计算机指令后，计算机运行上述任一项所述基于视觉信息的相机检测方法。At the same time, the present invention also discloses a computer-readable storage medium, the storage medium stores computer instructions, and after the computer reads the computer instructions in the storage medium, the computer executes any of the above-mentioned visual information-based camera detection methods .
综上可知，本申请的技术方案通过计算图像亮度及清晰度、潜在遮挡的几何分布及机器学习聚类多层筛选，对相机遮挡和强光或弱光环境进行检测并进行真实性和误判检查，可以解决视觉移动机器人算法容易受到视觉传感器采集到图像的质量及周围环境的影响导致算法失效或结果错误的问题，提高视觉移动机器人算法的精确度和鲁棒性。To sum up, the technical solution of the present application detects camera occlusion and strong light or weak light environment, and conducts authenticity and misjudgment by calculating image brightness and clarity, geometric distribution of potential occlusion, and multi-layer screening of machine learning clustering. The inspection can solve the problem that the visual mobile robot algorithm is easily affected by the quality of the image collected by the visual sensor and the surrounding environment, resulting in algorithm failure or wrong results, and improve the accuracy and robustness of the visual mobile robot algorithm.
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。The above descriptions are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention shall be included in the protection of the present invention. within the range.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111492771.3ACN114170185A (en) | 2021-12-08 | 2021-12-08 | Camera detection method and system based on visual information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111492771.3ACN114170185A (en) | 2021-12-08 | 2021-12-08 | Camera detection method and system based on visual information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114170185Atrue CN114170185A (en) | 2022-03-11 |
Family
ID=80484517
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111492771.3APendingCN114170185A (en) | 2021-12-08 | 2021-12-08 | Camera detection method and system based on visual information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114170185A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011182147A (en)* | 2010-03-01 | 2011-09-15 | Fujitsu Semiconductor Ltd | Photographing scene determining device |
| CN105975980A (en)* | 2016-04-27 | 2016-09-28 | 百度在线网络技术（北京）有限公司 | Method of monitoring image mark quality and apparatus thereof |
| CN111860120A (en)* | 2020-06-03 | 2020-10-30 | 江西江铃集团新能源汽车有限公司 | Automatic shielding detection method and device for vehicle-mounted camera |
| CN111932596A (en)* | 2020-09-27 | 2020-11-13 | 深圳佑驾创新科技有限公司 | Method, device and equipment for detecting camera occlusion area and storage medium |
- 2021
- 2021-12-08CNCN202111492771.3Apatent/CN114170185A/enactivePending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011182147A (en)* | 2010-03-01 | 2011-09-15 | Fujitsu Semiconductor Ltd | Photographing scene determining device |
| CN105975980A (en)* | 2016-04-27 | 2016-09-28 | 百度在线网络技术（北京）有限公司 | Method of monitoring image mark quality and apparatus thereof |
| CN111860120A (en)* | 2020-06-03 | 2020-10-30 | 江西江铃集团新能源汽车有限公司 | Automatic shielding detection method and device for vehicle-mounted camera |
| CN111932596A (en)* | 2020-09-27 | 2020-11-13 | 深圳佑驾创新科技有限公司 | Method, device and equipment for detecting camera occlusion area and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| S. ZHAO等: "MaskFlownet: Asymmetric Feature Matching With Learnable Occlusion Mask", 《2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)》, 19 June 2020 (2020-06-19), pages 6277 - 6286* |
| 刘磊等: "基于遮挡检测的尺度自适应相关滤波跟踪", 《计算机系统应用》, vol. 27, no. 10, 25 October 2018 (2018-10-25), pages 285 - 290* |
| 邓哲也等: "用于表格事实检测的图神经网络模型", 《软件学报》, vol. 32, no. 03, 11 March 2021 (2021-03-11), pages 753 - 762* |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110060237B (en) | Fault detection method, device, equipment and system | |
| CN109816644B (en) | An automatic detection system for bearing defects based on multi-angle light source images | |
| CN111612781A (en) | A screen defect detection method, device and head-mounted display device | |
| CN111898610B (en) | Card unfilled corner detection method, device, computer equipment and storage medium | |
| EP1808018B1 (en) | Image sensor annotation method and apparatus | |
| CN116777877A (en) | Circuit board defect detection method, device, computer equipment and storage medium | |
| CN114359161A (en) | A defect detection method, apparatus, equipment and storage medium | |
| CN112801963B (en) | Video image occlusion detection method and system | |
| CN114627113B (en) | A printed circuit board defect detection method, system, device and medium | |
| CN114820621B (en) | Bolt loss defect detection method, system and device | |
| Ma et al. | UP-CrackNet: Unsupervised pixel-wise road crack detection via adversarial image restoration | |
| CN120235871A (en) | Intelligent detection system and method of drug particles based on image processing | |
| CN117745552A (en) | Self-adaptive image enhancement method and device and electronic equipment | |
| JP4533836B2 (en) | Fluctuating region detection apparatus and method | |
| CN116071692A (en) | Morphological image processing-based water gauge water level identification method and system | |
| CN119991544A (en) | Intelligent classification method and system for transmission line tension clamp defects based on two-dimensional images | |
| CN110910497A (en) | Method and system for realizing augmented reality map | |
| CN111598943B (en) | Book in-place detection method, device and equipment based on book auxiliary reading equipment | |
| CN118365608A (en) | Camera content anomaly detection method, device, equipment and storage medium | |
| CN114170185A (en) | Camera detection method and system based on visual information | |
| CN115937263B (en) | Vision-based target tracking method, system, electronic device and storage medium | |
| CN116912167A (en) | Transmission line inspection method, system and controller | |
| CN116416232A (en) | Target detection method, device, electronic device, and computer-readable storage medium | |
| CN116309780A (en) | Water gauge water level identification method based on target detection | |
| CN107016692B (en) | A moving shadow detection method based on computer vision |
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 |