技术领域technical field
本发明属于计算机视觉处理技术领域，具体涉及一种自动泊车停车位检测与识别系统，可用于驾驶员寻找泊车位置。The invention belongs to the technical field of computer vision processing, and in particular relates to an automatic parking parking space detection and recognition system, which can be used for drivers to find parking positions.
背景技术Background technique
随着汽车保有量的快速上升，在城市场景中“停车难”的问题表现的越来越突出。驾驶员在泊车时，由于受到视野和车位狭小等客观条件的限制，以及技术和心理上的影响，极易发生擦碰事故，带来不必要的损失。一个有效的停车辅助系统，能帮助驾驶员快速、安全的完成泊车操作，自动泊车系统能够不借助驾驶员的操控自动完成安全、准确的泊车入位。With the rapid increase of car ownership, the problem of "difficult parking" in urban scenes has become more and more prominent. When the driver is parking, due to the limitation of objective conditions such as narrow vision and parking spaces, as well as technical and psychological influences, it is very easy to have a collision accident and cause unnecessary losses. An effective parking assistance system can help the driver to complete the parking operation quickly and safely, and the automatic parking system can automatically complete the safe and accurate parking position without the driver's control.
目前，市场上常用的自动泊车停车位检测与识别方法是基于超声波雷达的方法。然而，基于超声波雷达的方法要求目标停车位前后均停有车辆方可实现车位检测，且超声波雷达检测方法有着检测范围小、存在盲区等的缺点。针对这一问题，将超声波与机器视觉相结合的基于视觉的泊车辅助系统无疑是未来泊车辅助系统的发展方向。但是当前基于视觉的停车位检测与识别方法只针对普通场景下地面停车位标记线的检测，而忽略了目标停车位存在障碍物、阴影、标记线不全等复杂情况下的停车位检测问题，所以如何在较为复杂的场景下能更为准确、可靠地检测停车位是解决当前交通管理的急需问题。At present, the commonly used automatic parking space detection and recognition methods on the market are based on ultrasonic radar. However, the method based on ultrasonic radar requires vehicles to be parked before and after the target parking space to realize parking space detection, and the ultrasonic radar detection method has the disadvantages of small detection range and blind spots. In response to this problem, the vision-based parking assistance system that combines ultrasonic and machine vision is undoubtedly the development direction of the future parking assistance system. However, the current vision-based parking space detection and recognition methods only aim at the detection of marking lines of ground parking spaces in ordinary scenes, and ignore the parking space detection problems in complex situations such as obstacles, shadows, and incomplete marking lines in the target parking space. How to detect parking spaces more accurately and reliably in more complex scenarios is an urgent problem to solve the current traffic management.
发明内容Contents of the invention
本发明的目的在于针对上述现有技术的不足，提出基于全景视觉辅助系统的自动泊车停车位检测与识别系统，以提高驾驶员在复杂的场景下检测泊车位置的准确性和可靠性。The object of the present invention is to address the deficiencies of the above-mentioned prior art, and propose an automatic parking detection and recognition system based on a panoramic visual assistance system, so as to improve the accuracy and reliability of the driver's detection of the parking position in a complex scene.
实现本发明目的的技术思路是：通过安装在车身周围的四个广角摄像头，生成车辆周围环境全景图像，再结合计算机视觉算法对车身周围停车位进行检测与识别，以识别存在障碍物、阴影、标记线不全这些特殊情况下，实现高效、准确的停车位检测与识别。The technical thought of realizing the purpose of the present invention is: through four wide-angle cameras installed around the vehicle body, generate the panoramic image of the surrounding environment of the vehicle, and then combine the computer vision algorithm to detect and identify the parking spaces around the vehicle body to identify obstacles, shadows, Efficient and accurate detection and recognition of parking spaces can be achieved under special circumstances such as incomplete marking lines.
根据上述思路，本发明基于全景视觉辅助系统的自动泊车停车位检测与识别系统，包括：According to the above thinking, the present invention is based on the automatic parking detection and recognition system of the panoramic visual assistance system, including:
全景拍摄单元，用于采集车身周围图像，并通过摄像机标定、畸变矫正、俯视图变换和图像拼接生成无缝拼接的360°俯视图，传送至嵌入式单元；The panoramic shooting unit is used to collect images around the vehicle body, and generate a seamless 360° top view through camera calibration, distortion correction, top view transformation and image stitching, and transmit it to the embedded unit;
嵌入式单元，用于对全景拍摄模块采集的俯视图进行处理，完成停车位的检测并判定停车位是否被占用，并将检测结果发送至车载电子控制单元ECU和车载中控显示屏；The embedded unit is used to process the top view collected by the panoramic shooting module, complete the detection of the parking space and determine whether the parking space is occupied, and send the detection result to the vehicle electronic control unit ECU and the vehicle central control display;
车载中控显示屏，用于显示嵌入式单元所传送的停车位检测结果；The on-board central control display screen is used to display the parking space detection results transmitted by the embedded unit;
其特征在于：所述嵌入式单元包括：It is characterized in that: the embedded unit includes:
接口模块，用于将嵌入式单元检测结果发送至车载电子控制单元ECU；The interface module is used to send the detection result of the embedded unit to the vehicle electronic control unit ECU;
图像处理模块，其包括检测子模块和识别子模块，An image processing module, which includes a detection submodule and a recognition submodule,
该检测子模块，对无缝拼接的360°俯视图先进行图像预处理，再运用线段检测算法提取出俯视图中的停车位标记线，并计算停车位标记线的周长，根据周长判定停车位标记线是否完整：如不完整则运用图像分割算法将停车位标记线补全，再判定是否为可选停车位；The detection sub-module first performs image preprocessing on the seamlessly spliced 360° top view, and then uses the line segment detection algorithm to extract the parking space marking line in the top view, and calculates the perimeter of the parking space marking line, and judges the parking space according to the perimeter Whether the marking line is complete: if it is incomplete, use the image segmentation algorithm to complete the marking line of the parking space, and then determine whether it is an optional parking space;
该识别子模块，对检测子模块检测出的目标停车位，通过计算停车位内部灰度变化差异值，判定可选停车位内部是否存在阴影或障碍物，如存在障碍物再计算障碍物高度，从而判定可选停车位是否被占用；The recognition sub-module, for the target parking space detected by the detection sub-module, determines whether there is a shadow or an obstacle inside the optional parking space by calculating the difference value of the gray scale change inside the parking space, and calculates the height of the obstacle if there is an obstacle, To determine whether the optional parking space is occupied;
图像显示子模块，用于将嵌入式单元可选停车位检测结果在车载中控显示屏中显示。The image display sub-module is used to display the detection result of the optional parking space of the embedded unit on the vehicle central control display screen.
本发明具有如下优点或有益效果：The present invention has following advantage or beneficial effect:
本发明由于检测过程不依赖相邻汽车的停放姿势而仅依赖于车位停车线，在全景视觉辅助系统中，先利用LSD线段检测算法提取出俯视图中的停车位标记线，再通过计算停车位标记线各线段所围成的周长，提高了在停车位标记线不全的情况下停车位的检测准确度；此外由于本发明通过计算图像灰度差异值信息，先判定是否存在阴影或障碍物，如存在障碍物再判定障碍物高度，从而确定可选停车位是否被占用，有效解决了停车位存在障碍物、阴影情况下的停车位检测问题，实现了高效、准确的停车位检测与识别功能，为后续自动泊车中路径规划和路径跟踪控制提供了精度和稳定性的保障。Since the detection process of the present invention does not depend on the parking posture of adjacent cars but only on the parking space and parking line, in the panoramic vision assistance system, the LSD line segment detection algorithm is first used to extract the parking space marking line in the top view, and then by calculating the parking space mark The perimeter surrounded by each line segment of the line improves the detection accuracy of the parking space when the parking space marking line is incomplete; in addition, because the present invention first determines whether there is a shadow or an obstacle by calculating the image gray level difference value information, If there is an obstacle, then determine the height of the obstacle, so as to determine whether the optional parking space is occupied, effectively solve the problem of parking space detection when there are obstacles and shadows in the parking space, and realize efficient and accurate parking space detection and recognition functions , providing accuracy and stability guarantee for path planning and path tracking control in subsequent automatic parking.
以下结合附图本发明更加充分的描述本发明的实施例。然而，所附附图仅用于说明和阐述，并不构成对本发明范围的限制。Embodiments of the present invention will be described more fully below in conjunction with the accompanying drawings. However, the accompanying drawings are for illustration and illustration only, and do not limit the scope of the present invention.
附图说明Description of drawings
图1为本发明中自动泊车停车位检测与识别系统的结构示意图；Fig. 1 is the structural representation of automatic parking parking space detection and recognition system in the present invention;
图2为本发明中检测子模块进行停车位检测的过程图；Fig. 2 is the process diagram that detection sub-module carries out parking space detection among the present invention;
图3为本发明中识别子模块进行停车位识别的过程图。Fig. 3 is a process diagram of identifying a parking space by the identification sub-module in the present invention.
具体实施方式Detailed ways
一、技术原理1. Technical principles
本发明包括自动泊车停车位的检测和自动泊车停车位的识别，其中：The invention includes the detection of automatic parking spaces and the recognition of automatic parking spaces, wherein:
(一)自动泊车停车位的检测(1) Detection of automatic parking parking spaces
自动泊车停车位的检测包括图像预处理、停车标记线检测、停车标记线周长计算三个部分。The detection of automatic parking parking spaces includes three parts: image preprocessing, detection of parking marking lines, and calculation of the perimeter of parking marking lines.
所述图像预处理，包括图像灰度化和图像边缘检测，图像灰度化是指将摄像头采集的彩色图像转化为灰度图像，边缘检测是指检测识别出图像中亮度变化剧烈的像素点构成的集合，图像预处理过程主要用于减少图像中的噪声和外界的干扰，以简化后续处理过程。The image preprocessing includes image grayscale and image edge detection. Image grayscale refers to converting the color image collected by the camera into a grayscale image, and edge detection refers to detecting and identifying the composition of pixels with sharp brightness changes in the image. The image preprocessing process is mainly used to reduce the noise and external interference in the image to simplify the subsequent processing process.
所述停车标记线检测，是在前期边缘检测基础上，运用LSD线段检测算法对停车位中的停车标记线进行提取，该算法通过对图像局部分析，得出直线的像素点集，再通过假设参数进行验证求解，将像素点集与误差控制集合合并，进行自适应控制误检的数量，进而提高检测的准确性，是一种可以在线性时间内提取出亚像素级别图像直线特征的算法，与传统的Hough变换进行直线检测的方法比较LSD算法在检测直线段方面，检测精度与计算效率得到了一个很好的平衡。The parking marking line detection is based on the edge detection in the early stage, using the LSD line segment detection algorithm to extract the parking marking line in the parking space. The parameters are verified and solved, the pixel point set is combined with the error control set, and the number of false detections is adaptively controlled, thereby improving the accuracy of detection. It is an algorithm that can extract straight line features of sub-pixel images in linear time. Compared with the traditional Hough transform method for line detection, the LSD algorithm has a good balance between detection accuracy and calculation efficiency in the detection of line segments.
根据国标GB50067-2014，在实际场景中，停车位的宽度约为2.5米，停车位标记线的宽度约为10厘米。将平行线之间的物理距离设置为w1，范围为0.1米至0.15厘米，两个平行线段对之间的物理距离设置为w2，范围为2.3至2.8米。在边缘图像中检测出满足平行线之间的物理距离为w1且两个平行线段对之间的物理距离为w2的两个平行线段对，完成停车标记线检测。According to the national standard GB50067-2014, in the actual scene, the width of the parking space is about 2.5 meters, and the width of the parking space marking line is about 10 cm. Set the physical distance between parallel lines as w1 , ranging from 0.1 meters to 0.15 centimeters, and set the physical distance between two pairs of parallel line segments as w2 , ranging from 2.3 to 2.8 meters. In the edge image, two parallel line segment pairs satisfying that the physical distance between parallel lines is w1 and the physical distance between two parallel line segment pairs is w2 are detected to complete the detection of parking marking lines.
所述停车标记线周长计算，是针对目标停车位存在标记线不全情况下的停车位准确检测，需利用全景拍摄单元中摄像机标定所得到的摄像机内外参数及图像坐标系与世界坐标系之间的映射关系，计算出停车标记线周长，通过设置停车线周长高阈值和平均阈值，将计算出的停车标记线周长和设置的阈值进行比较，来判定停车位标记是否完全：若判定停车标记线不完全，则利用图像分割技术将缺失的停车线补全，形成闭合矩形，完成车位检测。The calculation of the perimeter of the parking marking line is aimed at the accurate detection of the parking space when the marking line is incomplete in the target parking space. According to the mapping relationship, the perimeter of the parking marking line is calculated. By setting the high threshold and the average threshold of the parking line perimeter, and comparing the calculated perimeter of the parking marking line with the set threshold, it is determined whether the parking space mark is complete: if the parking mark is determined If the parking line is incomplete, image segmentation technology is used to complete the missing parking line to form a closed rectangle to complete the parking space detection.
(二)自动泊车停车位的识别(2) Identification of automatic parking spaces
自动泊车停车位的识别，包括停车标记线内部区域灰度差异值计算和停车标记线内部障碍物高度计算两部分。The recognition of automatic parking parking spaces includes two parts: the calculation of the gray difference value of the inner area of the parking marking line and the calculation of the height of obstacles inside the parking marking line.
所述停车标记线内部区域灰度差异值计算，是由于停车位被占用时和未被占用时，停车标记线内部区域灰度值变化差异具有明显不同，需先计算出停车位标记线内部灰度变化差异值，再通过设置车位未被占用时灰度变化差异平均阈值，将停车位区域内部灰度变化差异值与车位未被占用时灰度变化差异平均阈值进行比较，以初步判定停车位内部是否被占用。The calculation of the gray difference value of the inner area of the parking marking line is due to the obvious difference in the gray value change difference of the inner area of the parking marking line when the parking space is occupied and not occupied, and it is necessary to first calculate the gray value of the parking space marking line Then, by setting the average threshold value of the gray scale change difference when the parking space is not occupied, the gray scale change difference value inside the parking space area is compared with the average gray scale change difference threshold value when the parking space is not occupied, so as to preliminarily determine the parking space. Whether the interior is occupied.
所述停车标记线内部障碍物高度计算，是在初步判定停车位内部区域存在障碍物情况下，需进一步判断障碍物高度，来最终判定停车位是否被占用：The calculation of the height of the obstacle inside the parking marking line is based on the preliminary determination that there is an obstacle in the inner area of the parking space, and the height of the obstacle needs to be further judged to finally determine whether the parking space is occupied:
首先，利用图像分割算法对停车位区域内障碍物进行提取，用最小外接矩形将障碍物在二维图像中显示出来，障碍物的顶点就是二维图像中最小外接矩形与障碍物之间的切点；First, the image segmentation algorithm is used to extract the obstacles in the parking space area, and the obstacles are displayed in the two-dimensional image with the minimum circumscribed rectangle. The vertex of the obstacle is the tangent between the minimum circumscribed rectangle and the obstacle point;
然后，利用单目立体视觉算法计算出障碍物的顶点的三维坐标，从而计算出障碍物高度，通过设置车辆平均最小离地间隙阈值，将计算出障碍物高度与车辆平均最小离地间隙阈值进行比较，最终判定停车位是否被占用。Then, use the monocular stereo vision algorithm to calculate the three-dimensional coordinates of the vertex of the obstacle, thereby calculating the height of the obstacle. By setting the average minimum ground clearance threshold of the vehicle, the calculated obstacle height is compared with the average minimum ground clearance threshold of the vehicle. Comparison, finally determine whether the parking space is occupied.
二、系统结构2. System structure
以下结合附图1，对本项发明的实例进行阐述，但本发明的保护范围并不仅限于此。Below in conjunction with accompanying drawing 1, the example of the present invention is described, but the scope of protection of the present invention is not limited thereto.
参照图1，本发明基于全景视觉辅助系统的自动泊车停车位检测与识别系统，包括全景拍摄单元、车载中控大屏、嵌入式处理单元和电源单元。所述嵌入式处理单元连接全景拍摄单元并与车载中控大屏双向链接，电源单元连接嵌入式处理单元、全景拍摄单元，全景拍摄单元连接嵌入式处理单元。其中：Referring to Fig. 1, the present invention is based on the automatic parking parking space detection and recognition system of the panoramic vision assistance system, including a panoramic shooting unit, a vehicle-mounted central control large screen, an embedded processing unit and a power supply unit. The embedded processing unit is connected to the panoramic shooting unit and bidirectionally linked with the vehicle-mounted central control large screen, the power supply unit is connected to the embedded processing unit and the panoramic shooting unit, and the panoramic shooting unit is connected to the embedded processing unit. in:
全景拍摄单元，包括4个170°广角摄像头，前摄像头安装在车辆进气格栅车标下方，两侧摄像头安装车辆B柱上方，后摄像头安装在牌照架上方，将拍摄到车辆前后左右四副广角图像处理生成显示车辆周围场景的360°无缝拼接的俯视图，并发送至嵌入式处理单元；Panoramic shooting unit, including 4 170° wide-angle cameras, the front camera is installed under the vehicle grille logo, the cameras on both sides are installed above the B-pillar of the vehicle, and the rear camera is installed above the license plate frame, which will capture the front, rear, left, and right sides of the vehicle Wide-angle image processing generates a seamless 360° top view showing the scene around the vehicle and sends it to the embedded processing unit;
嵌入式单元，主要负责完成对无缝拼接的俯视图的停车位检测与识别，并将检测结果发送至车载电子控制单元ECU，为后续自动泊车中路径规划和路径跟踪控制提供依据，同时，将检测结果通过图像显示模块发送至车载中控显示屏，其包括接口模块、图像显示模块、图像处理模块；The embedded unit is mainly responsible for the detection and recognition of parking spaces in the seamless top view, and sends the detection results to the on-board electronic control unit ECU to provide a basis for path planning and path tracking control in subsequent automatic parking. The detection result is sent to the vehicle central control display through the image display module, which includes an interface module, an image display module, and an image processing module;
接口模块，用于分别与全景拍摄单元、车载电子控制单元ECU、车载中控大屏、电源单元进行连接；The interface module is used to connect with the panoramic shooting unit, the vehicle electronic control unit ECU, the vehicle central control large screen and the power supply unit respectively;
图像处理模块，用于完成停车位的检测与识别功能，包括检测子模块和识别子模块。The image processing module is used to complete the detection and recognition function of the parking space, including a detection sub-module and a recognition sub-module.
图像显示模块，用于将图像处理模块检测的结果发送至车载中控显示屏，为驾驶员显示出检测结果，以挑选目标停车位。The image display module is used to send the detection result of the image processing module to the vehicle central control display, and display the detection result for the driver to select the target parking space.
参照图2，所述图像处理模块中的检测子模块，包括图像预处理、停车标记线检测、停车标记线周长计算三个部分。其中：Referring to FIG. 2 , the detection sub-module in the image processing module includes three parts: image preprocessing, detection of parking marking lines, and calculation of perimeter length of parking marking lines. in:
图像预处理，包括图像灰度化和图像边缘检测，即先将彩色图像转化为灰度图像使图像灰度化，再采用Canny算子完成边缘检测，该Canny算子能够满足低错误率、高定位性、最小响应三个最优边缘检测评价标准；Image preprocessing, including image grayscale and image edge detection, that is, first convert the color image into a grayscale image to grayscale the image, and then use the Canny operator to complete the edge detection. The Canny operator can meet the requirements of low error rate, high Positioning and minimum response three optimal edge detection evaluation criteria;
停车标记线线段提取，是在图像预处理的基础上，运用LSD线段检测算法对边缘图像中的线段进行提取，在提取出的线段中，根据国标GB50067-2014，检测出满足平行线之间的物理距离为w1且两个平行线段对之间的物理距离为w2的两个平行线段对，完成停车标记线检测。The line segment extraction of the parking marking line is based on the image preprocessing, using the LSD line segment detection algorithm to extract the line segments in the edge image. Among the extracted line segments, according to the national standard GB50067-2014, it is detected to meet the requirements between parallel lines. Two parallel line segment pairs with a physical distance of w1 and a physical distance between the two parallel line segment pairs of w2 complete the detection of parking marking lines.
停车位标记线周长计算，是对于目标停车位存在标记线不全情况下的车位准确检测，利用全景拍摄单元中摄像头内外参数及停车标记线在图像坐标系和世界坐标系的映射关系，先计算出停车位标记线各线段所围成的周长P1，再设置停车线周长两个阈值，高阈值Ph和平均阈值Pm，将计算出停车位标记线各线段所围成的周长P1与设置的停车线周长两个阈值进行比较：The calculation of the perimeter of the parking space marking line is to accurately detect the parking space when the marking line is incomplete in the target parking space. Using the internal and external parameters of the camera in the panoramic shooting unit and the mapping relationship between the parking marking line in the image coordinate system and the world coordinate system, first calculate Out of the perimeter P1 enclosed by each line segment of the parking space marking line, and then set two thresholds for the parking line perimeter, the high threshold Ph and the average threshold Pm , the perimeter enclosed by each line segment of the parking space marking line will be calculated. The length P1 is compared with the two thresholds set for the perimeter of the stop line:
若停车线周长P1＞Ph，则判定为停车位标记线完全。If the circumference of the parking line P1 >Ph , it is determined that the parking space marking line is complete.
若停车线周长Ph＞P1＞Pm，则判定为停车位标记线不完全；If the circumference of the parking line Ph >P1 >Pm , it is determined that the parking space marking line is incomplete;
当判定为停车位标记线不完全时，利用图像分割技术将缺失的停车线补全，形成闭合矩形、完成车位检测。When it is determined that the marking line of the parking space is incomplete, the missing parking line is completed using image segmentation technology to form a closed rectangle and complete the detection of the parking space.
参照图3，所述图像处理模块中识别子模块，包括停车标记线内部区域灰度差异值和障碍物高度计算两个部分。其中：Referring to FIG. 3 , the recognition sub-module in the image processing module includes two parts, the gray difference value of the inner area of the parking marking line and the calculation of the obstacle height. in:
停车标记线内部区域灰度差异值计算，是根据停车标记线内部的灰度直方图，计算出停车标记线内部区域灰度差异值，即先设置车位未被占用时灰度变化差异平均阈值T，再将计算出的灰度差异值T1与未被占用时灰度变化差异平均阈值T进行比较：The calculation of the gray level difference value of the inner area of the parking marking line is based on the gray level histogram inside the parking marking line to calculate the gray level difference value of the inner area of the parking marking line, that is, first set the average gray level difference threshold T when the parking space is not occupied , and then compare the calculated gray-scale difference T1 with the average threshold T of gray-scale change difference when it is not occupied:
若灰度变化差异值T1＞T，则判定为停车预期内无障碍物；If the difference value of gray scale change T1 >T, it is judged that there is no obstacle within the parking expectation;
若灰度变化差异值T1＜T，则判定为停车预期内有障碍物；If the difference value of the gray scale change T1 <T, it is determined that there is an obstacle within the parking expectation;
停车位标记线内部障碍物高度计算，是先当判定为停车预期内有障碍物时，对停车位内部利用图像分割算法标记出障碍物轮廓，利用车辆位移产生的视觉差，利用立体视觉算法计算出障碍物高度H1，通过设置车辆最小离地间隙再将计算出的障碍物高度H1与车辆离地最小间隙平均阈值H进行比较：The calculation of the obstacle height inside the parking space marking line is to first use the image segmentation algorithm to mark the outline of the obstacle inside the parking space when it is determined that there is an obstacle in the parking space, and use the visual difference generated by the vehicle displacement to calculate the height using the stereo vision algorithm. Obtain the obstacle height H1 , and compare the calculated obstacle height H1 with the average threshold H of the minimum ground clearance of the vehicle by setting the minimum ground clearance of the vehicle:
若障碍物高度H1＜H，则判定为车位未被占用，If the obstacle height H1 <H, it is determined that the parking space is not occupied,
若障碍物高度H1＞H，则判定为车位被占用。If the obstacle height H1 >H, it is determined that the parking space is occupied.
当系统识别为车位未被占用时，将识别结果通过嵌入式单元接口模块发送至车载电子控制单元ECU，以完成后续的自动泊车中路径规划和路径跟踪控制，同时，将识别结果通过嵌入式单元图像显示模块发送至车载中控大屏，为驾驶员显示出检测结果，以挑选目标停车位。When the system recognizes that the parking space is not occupied, the recognition result is sent to the vehicle electronic control unit ECU through the embedded unit interface module to complete the subsequent path planning and path tracking control in automatic parking. At the same time, the recognition result is passed through the embedded The unit image display module sends it to the large screen of the vehicle's central control to display the detection results for the driver to select the target parking space.
当系统识别为车位被占用时，将识别结果发送至嵌入式单元图像处理模块检测子模块，重新进行新一轮的停车位检测。When the system recognizes that the parking space is occupied, the recognition result is sent to the detection sub-module of the embedded unit image processing module, and a new round of parking space detection is performed again.
以上描述仅是本发明的一个具体事例，并未构成对本发明的任何限制，显然对于本领域的专业人员来说，在了解了本发明内容和原理后，都可能在不背离本发明原理、结构的情况下，进行形式和细节上的各种修改和改变，但是这些基于本发明思想的修正和改变仍在本发明的权力要求保护范围之内。The above description is only a specific example of the present invention, and does not constitute any limitation to the present invention. Obviously, for those skilled in the art, after understanding the content and principle of the present invention, it is possible without departing from the principle and structure of the present invention. Various modifications and changes in form and details are made, but these modifications and changes based on the idea of the present invention are still within the protection scope of the claims of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710864999.8ACN107738612B (en) | 2017-09-22 | 2017-09-22 | Automatic parking parking space detection and recognition system based on panoramic vision assistance system |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710864999.8ACN107738612B (en) | 2017-09-22 | 2017-09-22 | Automatic parking parking space detection and recognition system based on panoramic vision assistance system |
| Publication Number | Publication Date |
|---|---|
| CN107738612Atrue CN107738612A (en) | 2018-02-27 |
| CN107738612B CN107738612B (en) | 2020-04-14 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710864999.8AActiveCN107738612B (en) | 2017-09-22 | 2017-09-22 | Automatic parking parking space detection and recognition system based on panoramic vision assistance system |
| Country | Link |
|---|---|
| CN (1) | CN107738612B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109278861A (en)* | 2018-07-05 | 2019-01-29 | 惠州市德赛西威汽车电子股份有限公司 | A kind of positive based reminding method of steering wheel return looked around based on automobile and system |
| CN109508682A (en)* | 2018-11-20 | 2019-03-22 | 成都通甲优博科技有限责任公司 | A kind of detection method on panorama parking stall |
| CN109631896A (en)* | 2018-07-23 | 2019-04-16 | 同济大学 | A kind of parking lot autonomous parking localization method based on vehicle vision and motion information |
| CN109649384A (en)* | 2019-02-15 | 2019-04-19 | 华域汽车系统股份有限公司 | A kind of parking assistance method |
| CN109685000A (en)* | 2018-12-21 | 2019-04-26 | 广州小鹏汽车科技有限公司 | A kind of method for detecting parking stalls and device of view-based access control model |
| CN109727481A (en)* | 2019-01-08 | 2019-05-07 | 北京优位智停科技有限公司 | A multi-eye parking space monitoring and detection device |
| CN109740521A (en)* | 2018-12-29 | 2019-05-10 | 百度在线网络技术（北京）有限公司 | The parking stall location determining method and device of automatic parking, electronic equipment and computer-readable medium |
| CN109754639A (en)* | 2019-02-28 | 2019-05-14 | 北京智行者科技有限公司 | It parks storage method and device |
| CN109934140A (en)* | 2019-03-01 | 2019-06-25 | 武汉光庭科技有限公司 | Automatic backing method for assisting in parking and system based on detection ground horizontal marking |
| CN110415555A (en)* | 2019-08-30 | 2019-11-05 | 的卢技术有限公司 | A kind of recognition methods of effective scribing line parking stall and system based on deep learning |
| CN110544386A (en)* | 2019-09-18 | 2019-12-06 | 奇瑞汽车股份有限公司 | parking space identification method and device and storage medium |
| CN110751850A (en)* | 2019-08-30 | 2020-02-04 | 的卢技术有限公司 | Parking space identification method and system based on deep neural network |
| CN110889974A (en)* | 2018-09-11 | 2020-03-17 | 广州汽车集团股份有限公司 | Intelligent parking space identification method and device and automobile |
| CN111071248A (en)* | 2019-12-24 | 2020-04-28 | 浙江合众新能源汽车有限公司 | Intelligent parking method based on electronic power-assisted brake |
| CN111429469A (en)* | 2019-04-17 | 2020-07-17 | 杭州海康威视数字技术股份有限公司 | Parking position determining method and device, electronic equipment and storage medium |
| CN111746503A (en)* | 2019-03-27 | 2020-10-09 | 上海欧菲智能车联科技有限公司 | Parking method, system, device, vehicle and computer readable storage medium |
| CN111968132A (en)* | 2020-07-28 | 2020-11-20 | 哈尔滨工业大学 | Panoramic vision-based relative pose calculation method for wireless charging alignment |
| CN112009461A (en)* | 2019-05-13 | 2020-12-01 | 上海博泰悦臻网络技术服务有限公司 | Parking assist method and parking assist system |
| CN112183413A (en)* | 2020-09-30 | 2021-01-05 | 东风柳州汽车有限公司 | Parking space detection method and device, storage medium and vehicle |
| CN112348845A (en)* | 2019-08-06 | 2021-02-09 | 黑芝麻智能科技(重庆)有限公司 | System and method for parking space detection and tracking |
| CN112389419A (en)* | 2019-08-12 | 2021-02-23 | 欧特明电子股份有限公司 | Method for identifying parking space and parking assistance system |
| CN112767425A (en)* | 2020-12-30 | 2021-05-07 | 智车优行科技（北京）有限公司 | Parking space detection method and device based on vision |
| CN113060125A (en)* | 2020-08-31 | 2021-07-02 | 惠州华阳通用电子有限公司 | Parking assisting method and device |
| CN113320474A (en)* | 2021-07-08 | 2021-08-31 | 长沙立中汽车设计开发股份有限公司 | Automatic parking method and device based on panoramic image and human-computer interaction |
| CN113554882A (en)* | 2021-07-20 | 2021-10-26 | 阿波罗智联(北京)科技有限公司 | Method, apparatus, device and storage medium for outputting information |
| US11541881B2 (en) | 2018-06-29 | 2023-01-03 | Byd Company Limited | Automatic parking method, device, system, and vehicle |
| CN115909795A (en)* | 2023-03-09 | 2023-04-04 | 斯润天朗（北京）科技有限公司 | Autonomous parking system and method based on parking lot cooperation |
| CN117237905A (en)* | 2022-06-06 | 2023-12-15 | 深圳联友科技有限公司 | Parking space detection model training method and electronic equipment |
| CN117261877A (en)* | 2023-08-18 | 2023-12-22 | 广州优保爱驾科技有限公司 | Self-correction image acquisition system and method based on vehicle appearance change |
| CN118015599A (en)* | 2024-04-09 | 2024-05-10 | 深圳市前海铼停科技有限公司 | Binocular parking identification method, system and processor based on image stitching |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112289070B (en)* | 2020-10-20 | 2021-12-07 | 广州小鹏自动驾驶科技有限公司 | Parking space detection method and device, vehicle and readable medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000134608A (en)* | 1998-10-22 | 2000-05-12 | Fujitsu Ten Ltd | Rear monitoring device for vehicle |
| CN102774324A (en)* | 2012-07-06 | 2012-11-14 | 广东好帮手电子科技股份有限公司 | Panoramic parking assist system and implementation method thereof |
| CN103171552A (en)* | 2011-12-23 | 2013-06-26 | 现代自动车株式会社 | AVM top view based parking support system |
| CN105539430A (en)* | 2015-12-29 | 2016-05-04 | 北京理工大学 | Intelligent man-car interaction parking method based on hand-held terminal |
| US20160129939A1 (en)* | 2013-07-04 | 2016-05-12 | Jaguar Land Rover Limited | Vehicle control system |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000134608A (en)* | 1998-10-22 | 2000-05-12 | Fujitsu Ten Ltd | Rear monitoring device for vehicle |
| CN103171552A (en)* | 2011-12-23 | 2013-06-26 | 现代自动车株式会社 | AVM top view based parking support system |
| CN102774324A (en)* | 2012-07-06 | 2012-11-14 | 广东好帮手电子科技股份有限公司 | Panoramic parking assist system and implementation method thereof |
| US20160129939A1 (en)* | 2013-07-04 | 2016-05-12 | Jaguar Land Rover Limited | Vehicle control system |
| CN105539430A (en)* | 2015-12-29 | 2016-05-04 | 北京理工大学 | Intelligent man-car interaction parking method based on hand-held terminal |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11541881B2 (en) | 2018-06-29 | 2023-01-03 | Byd Company Limited | Automatic parking method, device, system, and vehicle |
| CN109278861A (en)* | 2018-07-05 | 2019-01-29 | 惠州市德赛西威汽车电子股份有限公司 | A kind of positive based reminding method of steering wheel return looked around based on automobile and system |
| CN109631896A (en)* | 2018-07-23 | 2019-04-16 | 同济大学 | A kind of parking lot autonomous parking localization method based on vehicle vision and motion information |
| CN110889974B (en)* | 2018-09-11 | 2021-02-19 | 广州汽车集团股份有限公司 | A kind of intelligent parking space identification method and device, automobile |
| CN110889974A (en)* | 2018-09-11 | 2020-03-17 | 广州汽车集团股份有限公司 | Intelligent parking space identification method and device and automobile |
| CN109508682A (en)* | 2018-11-20 | 2019-03-22 | 成都通甲优博科技有限责任公司 | A kind of detection method on panorama parking stall |
| CN109685000A (en)* | 2018-12-21 | 2019-04-26 | 广州小鹏汽车科技有限公司 | A kind of method for detecting parking stalls and device of view-based access control model |
| WO2020124988A1 (en)* | 2018-12-21 | 2020-06-25 | 广州小鹏汽车科技有限公司 | Vision-based parking space detection method and device |
| CN109740521A (en)* | 2018-12-29 | 2019-05-10 | 百度在线网络技术（北京）有限公司 | The parking stall location determining method and device of automatic parking, electronic equipment and computer-readable medium |
| CN109727481A (en)* | 2019-01-08 | 2019-05-07 | 北京优位智停科技有限公司 | A multi-eye parking space monitoring and detection device |
| CN109649384B (en)* | 2019-02-15 | 2020-08-14 | 华域汽车系统股份有限公司 | Parking assisting method |
| CN109649384A (en)* | 2019-02-15 | 2019-04-19 | 华域汽车系统股份有限公司 | A kind of parking assistance method |
| CN109754639A (en)* | 2019-02-28 | 2019-05-14 | 北京智行者科技有限公司 | It parks storage method and device |
| CN109934140A (en)* | 2019-03-01 | 2019-06-25 | 武汉光庭科技有限公司 | Automatic backing method for assisting in parking and system based on detection ground horizontal marking |
| CN109934140B (en)* | 2019-03-01 | 2022-12-02 | 武汉光庭科技有限公司 | Automatic reversing auxiliary parking method and system based on detection of ground transverse marking |
| CN111746503A (en)* | 2019-03-27 | 2020-10-09 | 上海欧菲智能车联科技有限公司 | Parking method, system, device, vehicle and computer readable storage medium |
| CN111429469B (en)* | 2019-04-17 | 2023-11-03 | 杭州海康威视数字技术股份有限公司 | Berth position determining method and device, electronic equipment and storage medium |
| CN111429469A (en)* | 2019-04-17 | 2020-07-17 | 杭州海康威视数字技术股份有限公司 | Parking position determining method and device, electronic equipment and storage medium |
| CN112009461A (en)* | 2019-05-13 | 2020-12-01 | 上海博泰悦臻网络技术服务有限公司 | Parking assist method and parking assist system |
| CN112348845A (en)* | 2019-08-06 | 2021-02-09 | 黑芝麻智能科技(重庆)有限公司 | System and method for parking space detection and tracking |
| CN112348845B (en)* | 2019-08-06 | 2024-02-23 | 黑芝麻智能科技(重庆)有限公司 | System and method for parking space detection and tracking |
| CN112389419B (en)* | 2019-08-12 | 2022-04-08 | 欧特明电子股份有限公司 | Method for identifying parking space and parking assistance system |
| CN112389419A (en)* | 2019-08-12 | 2021-02-23 | 欧特明电子股份有限公司 | Method for identifying parking space and parking assistance system |
| CN110751850A (en)* | 2019-08-30 | 2020-02-04 | 的卢技术有限公司 | Parking space identification method and system based on deep neural network |
| CN110415555A (en)* | 2019-08-30 | 2019-11-05 | 的卢技术有限公司 | A kind of recognition methods of effective scribing line parking stall and system based on deep learning |
| CN110751850B (en)* | 2019-08-30 | 2023-03-07 | 的卢技术有限公司 | Parking space identification method and system based on deep neural network |
| CN110415555B (en)* | 2019-08-30 | 2020-09-29 | 的卢技术有限公司 | Effective lineation parking space identification method and system based on deep learning |
| CN110544386A (en)* | 2019-09-18 | 2019-12-06 | 奇瑞汽车股份有限公司 | parking space identification method and device and storage medium |
| CN111071248B (en)* | 2019-12-24 | 2021-11-05 | 浙江合众新能源汽车有限公司 | An Intelligent Parking Method Based on Electronic Power Brake |
| CN111071248A (en)* | 2019-12-24 | 2020-04-28 | 浙江合众新能源汽车有限公司 | Intelligent parking method based on electronic power-assisted brake |
| CN111968132A (en)* | 2020-07-28 | 2020-11-20 | 哈尔滨工业大学 | Panoramic vision-based relative pose calculation method for wireless charging alignment |
| CN113060125A (en)* | 2020-08-31 | 2021-07-02 | 惠州华阳通用电子有限公司 | Parking assisting method and device |
| CN113060125B (en)* | 2020-08-31 | 2022-08-30 | 惠州华阳通用电子有限公司 | Parking assisting method and device |
| CN112183413B (en)* | 2020-09-30 | 2024-05-14 | 东风柳州汽车有限公司 | Parking space detection method and device, storage medium and vehicle |
| CN112183413A (en)* | 2020-09-30 | 2021-01-05 | 东风柳州汽车有限公司 | Parking space detection method and device, storage medium and vehicle |
| CN112767425A (en)* | 2020-12-30 | 2021-05-07 | 智车优行科技（北京）有限公司 | Parking space detection method and device based on vision |
| CN113320474A (en)* | 2021-07-08 | 2021-08-31 | 长沙立中汽车设计开发股份有限公司 | Automatic parking method and device based on panoramic image and human-computer interaction |
| CN113554882A (en)* | 2021-07-20 | 2021-10-26 | 阿波罗智联(北京)科技有限公司 | Method, apparatus, device and storage medium for outputting information |
| CN117237905A (en)* | 2022-06-06 | 2023-12-15 | 深圳联友科技有限公司 | Parking space detection model training method and electronic equipment |
| CN115909795A (en)* | 2023-03-09 | 2023-04-04 | 斯润天朗（北京）科技有限公司 | Autonomous parking system and method based on parking lot cooperation |
| CN117261877A (en)* | 2023-08-18 | 2023-12-22 | 广州优保爱驾科技有限公司 | Self-correction image acquisition system and method based on vehicle appearance change |
| CN117261877B (en)* | 2023-08-18 | 2024-05-14 | 广州优保爱驾科技有限公司 | Self-correction image acquisition system and method based on vehicle appearance change |
| CN118015599A (en)* | 2024-04-09 | 2024-05-10 | 深圳市前海铼停科技有限公司 | Binocular parking identification method, system and processor based on image stitching |
| Publication number | Publication date |
|---|---|
| CN107738612B (en) | 2020-04-14 |
| Publication | Publication Date | Title |
|---|---|---|
| CN107738612B (en) | Automatic parking parking space detection and recognition system based on panoramic vision assistance system | |
| CN112215306B (en) | Target detection method based on fusion of monocular vision and millimeter wave radar | |
| CN108759667B (en) | Distance measurement method of front vehicle based on monocular vision and image segmentation under vehicle camera | |
| CN106908783B (en) | Based on obstacle detection method combined of multi-sensor information | |
| CN106650708B (en) | Automatic driving obstacle vision detection method and system | |
| CN102867414B (en) | Vehicle queue length measurement method based on PTZ (Pan/Tilt/Zoom) camera fast calibration | |
| US9013286B2 (en) | Driver assistance system for displaying surroundings of a vehicle | |
| CN104574376B (en) | Avoiding collision based on binocular vision and laser radar joint verification in hustle traffic | |
| CN109435942A (en) | A kind of parking stall line parking stall recognition methods and device based on information fusion | |
| CN106128115B (en) | Fusion method for detecting road traffic information based on double cameras | |
| JP4930046B2 (en) | Road surface discrimination method and road surface discrimination device | |
| CN104881645B (en) | The vehicle front mesh object detection method of feature based point mutual information and optical flow method | |
| CN108805065A (en) | One kind being based on the improved method for detecting lane lines of geometric properties | |
| CN107392103A (en) | The detection method and device of road surface lane line, electronic equipment | |
| CN103600707A (en) | Parking position detecting device and method of intelligent parking system | |
| CN104063882B (en) | Vehicle video speed measuring method based on binocular camera | |
| US11900696B2 (en) | Method and system for processing a plurality of images so as to detect lanes on a road | |
| CN107256636A (en) | A kind of traffic flow acquisition methods for merging laser scanning and video technique | |
| CN105059190B (en) | The automobile door opening collision warning device and method of view-based access control model | |
| CN102303563B (en) | Front vehicle collision early warning system and method | |
| CN106802144A (en) | A kind of vehicle distance measurement method based on monocular vision and car plate | |
| CN111681283B (en) | Monocular stereoscopic vision-based relative pose calculation method applied to wireless charging alignment | |
| CN107229906A (en) | A kind of automobile overtaking's method for early warning based on units of variance model algorithm | |
| CN107122740A (en) | A kind of vehicle type recognition device and method based on twin camera | |
| CN111968132A (en) | Panoramic vision-based relative pose calculation method for wireless charging alignment |
| 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |