Obstacle Detection Method, System and Electronic Device Based on Optical Flow Feature Fusion

Abstract

The present invention provides an obstacle detection method, system and electronic device based on optical flow feature fusion. The method includes: collecting consecutive images during the parking process and preprocessing the collected images; dividing the preprocessed images into multiple grids, using an optical flow tracking algorithm to track the position changes of each grid in adjacent frames, obtaining the grids containing moving objects, fusing the grids containing moving objects to obtain the range of the moving object; dividing the range of the moving object according to a preset aspect ratio threshold, calculating the texture for each region respectively, and obtaining the region where the moving object is located. The present invention solves the problem that the detection of moving objects during automatic parking depends on ultrasonic radars and AI detection algorithms. Only by using the pictures recorded by the vehicle's own camera, it can be calculated whether there are moving objects in the pictures, with low cost, low hardware requirements, small computational amount, high speed and high detection accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent driving, and more specifically, to an obstacle detection method, system and electronic device based on optical flow feature fusion. Background Art

[0002] With the development of intelligent driving, automatic parking has become a configuration selected by more and more vehicle models. The purpose of automatic parking is to improve the convenience of parking, provide a safe, comfortable and fast parking service for drivers, and reduce the difficulty of drivers' parking. In the process of automatic parking, the detection of moving objects by a pure vision algorithm mainly relies on the change of optical flow features in the image to find the approximate range of the moving object, and then finds the position of the moving object through fine feature discrimination.

[0003] At present, the detection of moving objects in the process of automatic parking mainly relies on ultrasonic radars and AI detection algorithms. However, the equipment and installation costs of ultrasonic radars are relatively high, and the AI detection algorithm first finds suspicious targets and then performs position screening and matching, with a huge amount of calculation and high requirements for hardware, making it equally difficult to reduce costs. Summary of the Invention

[0004] In view of the technical problems existing in the prior art, the present invention provides an obstacle detection method, system and electronic device based on optical flow feature fusion, which solves the problem that the detection of moving objects in the process of automatic parking relies on relatively high-cost ultrasonic radars and AI detection algorithms.

[0005] According to a first aspect of the present invention, there is provided an obstacle detection method based on optical flow feature fusion, including:

[0006] Collecting consecutive images during the parking process and preprocessing the collected images;

[0007] Dividing the preprocessed images into multiple grids, using an optical flow tracking algorithm to track the position changes of each grid in adjacent frames, obtaining grids containing moving objects, and fusing the grids containing moving objects to obtain the range of the moving object;

[0008] According to a preset aspect ratio threshold, dividing the range of the moving object into regions, and respectively performing texture calculation on each region to obtain the region where the moving object is located.

[0009] On the basis of the above technical solutions, the present invention can also be improved as follows.

[0010] Optionally, the preprocessing of the collected images includes:

[0011] According to the current vehicle speed, thinning the consecutive frame images collected during the parking process to obtain a set of consecutive images with an interval of n frames, where n is a natural number greater than 1.

[0012] Optionally, dividing the preprocessed image into multiple grids and using an optical flow tracking algorithm to track the position changes of each grid in adjacent frames to obtain grids containing moving objects includes:

[0013] Dividing each frame of the preprocessed image into multiple grids respectively, and obtaining the coordinates of the center point of each grid;

[0014] Tracking the center point from the previous frame image to the next frame image using an optical flow tracking algorithm to obtain corresponding tracking points; performing anti-tracking on the tracking points from the next frame image to the previous frame image using an optical flow tracking algorithm to obtain the coordinates of the corresponding anti-tracking points;

[0015] Judging whether the center point coincides with its corresponding anti-tracking point according to the coordinate values: if the judgment result is coincidence, it is determined that the grid where the current center point is located moves; if the judgment result is non-coincidence, it is determined that the grid where the current center point is located is stationary.

[0016] Optionally, during the process of dividing the image into multiple grids, first exclude the edge area of the image according to the number of pixels, and then divide the remaining area into multiple grids.

[0017] Optionally, fusing the grids containing moving objects to obtain the range of the moving object includes:

[0018] Performing dilation and erosion operations on the grids containing moving objects, then screening the grids containing moving objects again, and fusing the newly obtained grids containing moving objects to obtain the range of the moving object.

[0019] Optionally, dividing the range of the moving object according to a preset aspect ratio threshold, calculating the texture of each area respectively, and obtaining the area where the moving object is located includes:

[0020] Calculating and judging whether the range of the moving object conforms to the aspect ratio of a normal moving object according to a preset first aspect ratio threshold;

[0021] If the range of the moving object conforms to the aspect ratio of a normal moving object, taking the range of the moving object as the area where the moving object is located;

[0022] If the range of the moving object does not conform to the aspect ratio of a normal moving object, dividing the range of the moving object into k areas according to a preset second aspect ratio threshold range, where k is a natural number greater than 1; calculating the horizontal texture, vertical texture and oblique texture of each divided area respectively, screening out the area that most conforms to the aspect ratio of a normal moving object according to the texture calculation results of each area, and taking this area as the area where the moving object is located.

[0023] Optionally, the second aspect ratio threshold range includes multiple aspect ratio threshold ranges, and each aspect ratio threshold range corresponds to a number of area divisions respectively.

[0024] According to a second aspect of the present invention, there is provided an obstacle detection system based on optical flow feature fusion, including:

[0025] An acquisition and preprocessing module, configured to acquire consecutive images during the parking process and preprocess the acquired images;

[0026] An optical flow tracking detection module, configured to divide the preprocessed image into multiple grids, use the optical flow tracking algorithm to track the position change of each grid in adjacent frames, obtain the grids containing moving objects, and fuse the grids containing moving objects to obtain the range of moving objects;

[0027] A fine feature recognition module, configured to divide the range of the moving object into regions according to a preset aspect ratio threshold, calculate the texture of each region respectively, and obtain the region where the moving object is located.

[0028] According to a third aspect of the present invention, there is provided an electronic device, including a memory and a processor, and the processor is configured to implement the steps of the obstacle detection method based on optical flow feature fusion when executing a computer management program stored in the memory.

[0029] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium, on which a computer management program is stored, and the computer management program implements the steps of the obstacle detection method based on optical flow feature fusion when executed by a processor.

[0030] An obstacle detection method, system, electronic device and storage medium based on optical flow feature fusion provided by the present invention solve the problem that the detection of moving objects during automatic parking depends on ultrasonic radars and AI detection algorithms. Only by using the pictures recorded by the vehicle's own camera, it can be calculated whether there are moving objects in the pictures, with low cost, low hardware requirements, small calculation amount, high speed and high detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a flowchart of the obstacle detection method based on optical flow feature fusion provided by the present invention;

[0032] Figure 2 It is a flowchart of optical flow tracking and anti-tracking of the method provided by the present invention;

[0033] Figure 3 It is a schematic diagram of grid division of an image in an embodiment;

[0034] Figure 4 It is a schematic diagram of a grid containing a moving object in an embodiment;

[0035] Figure 5 Schematic diagram of the area of the moving object after fusion in the embodiment;

[0036] Figure 6 Schematic diagram of the area conforming to the aspect ratio of the normal moving object in the embodiment;

[0037] Figure 7 Schematic diagram of dividing the range of the moving object into three areas in the embodiment;

[0038] Figure 8 Schematic diagram of dividing the range of the moving object into two areas in the embodiment;

[0039] Figure 9 Schematic diagram of screening out the area that best conforms to the moving target in the embodiment Figure 1 ;

[0040] Figure 10 Schematic diagram of screening out the area that best conforms to the moving target in the embodiment Figure 2 ;

[0041] Figure 11 Structural diagram of an obstacle detection system based on optical flow feature fusion provided by the present invention;

[0042] Figure 12 Schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0043] Figure 13 Schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. Detailed implementation manners

[0044] The following combines the accompanying drawings and embodiments to further describe in detail the specific implementation manners of the present invention. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.

[0045] Figure 1 Flowchart of an obstacle detection method based on optical flow feature fusion provided by the present invention, as Figure 1 shown, the method includes:

[0046] 101. Collect continuous images during the parking process and preprocess the collected images;

[0047] 102. Divide the preprocessed image into multiple grids, use the optical flow tracking algorithm to track the position changes of each grid in adjacent frames, obtain the grids containing moving objects, and fuse the grids containing moving objects to obtain the range of the moving object;

[0048] 103. Divide the range of the moving object into regions according to a preset aspect ratio threshold, calculate the texture of each region separately, and obtain the region where the moving object is located.

[0049] It can be understood that based on the deficiencies in the background technology, an obstacle detection method based on optical flow feature fusion is proposed in the embodiments of the present invention. This method solves the problem that the detection of moving objects during automatic parking depends on ultrasonic radars and AI detection algorithms. It is not necessary to install radars on the vehicle to detect moving objects. Only the pictures recorded by the vehicle's own camera (such as a monocular camera or a monocular fisheye camera) can be used to detect whether there are moving objects in the pictures through calculation, which has low cost and is convenient to use; it has low requirements for hardware, small calculation amount, high speed, and high detection accuracy.

[0050] In a possible embodiment, the preprocessing of the collected images includes:

[0051] According to the current vehicle speed, thin out the continuously captured frames of images during the parking process to obtain a set of consecutive images with an interval of n frames, where n is a natural number greater than 1.

[0052] It can be understood that since the vehicle speed during automatic parking is generally between 5 km / h and 10 km / h, that is, the vehicle moves a distance of 1.39 m to 2.78 m in 1 s. The monocular camera captures about 30 frames of pictures in 1 s, that is, the vehicle moves 0.046 m to 0.092 m between two adjacent frames. If the interval between two adjacent frames is 8 frames, the vehicle moves 0.368 m to 0.736 m. The pictures taken at this moving distance are most suitable for processing the optical flow change features on the pictures. Therefore, thin out the set of pictures captured by the camera, extract one picture every few frames of images to form a new set of images for subsequent optical flow tracking. After thinning out the images, the calculation amount of optical flow tracking is reduced on the basis of ensuring the subsequent detection accuracy, and the image processing speed is improved.

[0053] In a possible embodiment, as Figure 2 shown in the flowchart of, the step of dividing the preprocessed images into multiple grids and using the optical flow tracking algorithm to track the position changes of each grid in adjacent frames to obtain the grids containing moving objects includes steps 201, 202, and 203, where:

[0054] 201. Divide each frame of the preprocessed image into multiple grids respectively, and obtain the coordinates of the center point of each grid.

[0055] It can be understood that in this step, each frame of the preprocessed image is divided into multiple small grids respectively, and the center point of each grid is taken. For example, as Figure 3As shown in the image, the size of the picture captured by the camera is 1920*1080, and there will be problems with the loss of optical flow tracking at the edges of the picture. Therefore, in this step, first exclude the edge areas of the image according to the number of pixels, and then divide the remaining area into multiple grids. For example, the areas 100 pixels away from the left and right edges are regarded as edge areas and no optical flow tracking is performed; the areas 50 pixels away from the upper and lower edges are regarded as edge areas and no optical flow tracking is performed. Then divide the remaining area into 172*98 small grids, and record the center point coordinates of each small grid at the same time.

[0056] 202. Use the optical flow tracking algorithm to track the center point from the previous frame image to the next frame image to obtain the corresponding tracking points; use the optical flow tracking algorithm to perform reverse tracking from the next frame picture to the previous frame picture for the tracking points to obtain the coordinates of the corresponding reverse tracking points.

[0057] It can be understood that in this step, through optical flow tracking and reverse tracking, the coordinates of the reverse tracking points corresponding to the center points are obtained, and these reverse tracking point coordinates can be used as the basis for subsequent judgment on whether the objects in the grid move.

[0058] 203. Judge whether the center point coincides with its corresponding reverse tracking point according to the coordinate values: if the judgment result is coincidence, it is determined that the grid where the current center point is located moves; if the judgment result is non - coincidence, it is determined that the grid where the current center point is located is stationary.

[0059] It can be understood that in this step, the optical flow tracking results are verified. Through steps 201 and 202, the grid center points, the corresponding tracking points, and the corresponding reverse tracking points can be obtained. If the grid center point and the corresponding reverse tracking point basically coincide, it means that this center point is a stationary point, that is, this point does not move in the actual situation; on the contrary, if the grid center point and the corresponding reverse tracking point do not coincide, it means that this center point is moving in the actual situation. If the center point is stationary, it is estimated that the points in the entire small grid are also stationary; if the center point is moving, it is estimated that the points in the entire small grid are also moving. The grid schematic diagram containing the moving object obtained is as Figure 4 shown.

[0060] In a possible embodiment, the step of fusing the grids containing the moving object to obtain the range of the moving object includes:

[0061] After performing dilation and erosion operations on the grids containing the moving object, screen the grids containing the moving object again, and fuse the newly obtained grids containing the moving object to obtain the range of the moving object.

[0062] It can be understood that by performing step 102 or steps 201 - 203, it is possible to obtain which grids on the current image contain moving objects. Since optical flow tracking may have cases of loss and false detection, the obtained moving regions are sometimes not complete and ideal. Therefore, erosion and dilation operations need to be performed on these regions, followed by screening and fusion operations. As Figure 4 shown is the grid region containing the moving object extracted through optical flow tracking. As Figure 5 shown is the region of the fused moving object.

[0063] In a possible embodiment, the method of dividing the range of the moving object according to a preset aspect ratio threshold and calculating the texture for each region to obtain the region where the moving object is located includes:

[0064] According to a preset first aspect ratio threshold, calculate and determine whether the range of the moving object conforms to the aspect ratio of a normal moving object;

[0065] If the range of the moving object conforms to the aspect ratio of a normal moving object, then use the range of the moving object as the region where the moving object is located;

[0066] If the range of the moving object does not conform to the aspect ratio of a normal moving object, then divide the range of the moving object into k regions according to a preset second aspect ratio threshold range, where k is a natural number greater than 1; calculate the horizontal texture, vertical texture, and oblique texture for each divided region respectively, and screen out the region that most conforms to the aspect ratio of a normal moving object according to the texture calculation results of each region, and use this region as the region where the moving object is located.

[0067] It can be understood that in this embodiment, for the case where the range of the moving object is too large, fine feature recognition is performed. A first aspect ratio threshold range is set in advance to determine whether the obtained range of the moving object belongs to the range of a normal moving object. After comparing the aspect ratio of the range of the moving object obtained through step 102 with the first aspect ratio threshold range and determining that the aspect ratio of the obtained range of the moving object is that of a normal moving object, the region of the moving object can be directly obtained. As Figure 6 shown.

[0068] Through step 102, the approximate range of the moving object is obtained. However, sometimes due to sudden changes in vehicle speed or sudden changes in light, the range of the moving object detected through optical flow tracking may be too large and does not conform to the normal aspect ratio of the moving object. In this case, this range needs to be re - divided, and the characteristics of each divided region are calculated to determine whether they conform to the characteristics of the moving object.

[0069] As Figure 7 and Figure 8As shown, according to the aspect ratio of the moving object range, a second aspect ratio threshold range is set. According to the second aspect ratio threshold range, the moving object range is divided into two regions on the left and right or three regions on the left, middle, and right. Optionally, to increase the accuracy of the division, the second aspect ratio threshold range includes multiple aspect ratio threshold ranges, and each aspect ratio threshold range corresponds to a number of region divisions.

[0070] For example, when the aspect ratio range is from 1.2 to 2.4, the range is divided into two regions on the left and right; when the aspect ratio range is from 2.4 to 3.6, the range is divided into three regions on the left, middle, and right. Then, the horizontal texture, vertical texture, and oblique texture are calculated for each divided region, so as to screen out the region that best matches the moving target, as Figure 9 and Figure 10 shown, the dashed box represents the candidate region of the moving object, and the solid box represents the finally found region of the moving object.

[0071] Figure 11 The figure is a structural diagram of an obstacle detection system based on optical flow feature fusion provided by an embodiment of the present invention. As Figure 11 shown, an obstacle detection system based on optical flow feature fusion includes an acquisition and preprocessing module 1101, an optical flow tracking detection module 1102, and a fine feature recognition module 1106, where:

[0072] The acquisition and preprocessing module 1101 is used to collect continuous images during the parking process and preprocess the collected images;

[0073] The optical flow tracking detection module 1102 is used to divide the preprocessed image into multiple grids, use the optical flow tracking algorithm to track the position changes of each grid in adjacent frames, obtain the grids containing moving objects, and fuse the grids containing moving objects to obtain the moving object range;

[0074] The fine feature recognition module 1103 is used to divide the moving object range according to a preset aspect ratio threshold, calculate the texture for each region respectively, and obtain the region where the moving object is located.

[0075] It can be understood that an obstacle detection system based on optical flow feature fusion provided by the present invention corresponds to the obstacle detection method based on optical flow feature fusion provided in the foregoing embodiments. The related technical features of the obstacle detection system based on optical flow feature fusion can refer to the related technical features of the obstacle detection method based on optical flow feature fusion, which will not be elaborated here.

[0076] Please refer to Figure 12 , Figure 12 The figure is a schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. As Figure 12As shown in the figure, an embodiment of the present invention provides an electronic device, including a memory 1210, a processor 1220, and a computer program 1211 stored in the memory 1210 and executable on the processor 1220. When the processor 1220 executes the computer program 1211, the following steps are implemented:

[0077] Collect continuous images during the parking process and preprocess the collected images;

[0078] Divide the preprocessed image into multiple grids, use the optical flow tracking algorithm to track the position changes of each grid in adjacent frames, obtain the grids containing moving objects, and fuse the grids containing moving objects to obtain the range of the moving object;

[0079] According to a preset aspect ratio threshold, divide the range of the moving object into regions, calculate the texture of each region respectively, and obtain the region where the moving object is located.

[0080] Please refer to Figure 13 , Figure 13 which is a schematic diagram of an embodiment of a computer-readable storage medium provided by the present invention. As Figure 13 shown, this embodiment provides a computer-readable storage medium 1300, on which a computer program 1311 is stored. When the computer program 1311 is executed by a processor, the following steps are implemented:

[0081] Collect continuous images during the parking process and preprocess the collected images;

[0082] Divide the preprocessed image into multiple grids, use the optical flow tracking algorithm to track the position changes of each grid in adjacent frames, obtain the grids containing moving objects, and fuse the grids containing moving objects to obtain the range of the moving object;

[0083] According to a preset aspect ratio threshold, divide the range of the moving object into regions, calculate the texture of each region respectively, and obtain the region where the moving object is located.

[0084] An obstacle detection method, system, electronic device, and storage medium based on optical flow feature fusion provided by an embodiment of the present invention solve the problem that the detection of moving objects during automatic parking depends on ultrasonic radars and AI detection algorithms. Only by using the pictures recorded by the vehicle's own camera, it can be calculated whether there are moving objects in the pictures. It has low cost, low hardware requirements, small calculation amount, high speed, and high detection accuracy.

[0085] It should be noted that in the above embodiments, the descriptions of each embodiment have their own emphases. For parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0086] Those skilled in the art will understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0087] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce means for implementing the functions specified in one Figure 1 one or more flows and / or blocks Figure 1 or multiple flows and / or blocks and / or in one

[0088] or multiple blocks. Figure 1 These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one Figure 1 one or more flows and / or blocks

[0089] or multiple flows and / or blocks and / or in one Figure 1 one or more flows and / or blocks Figure 1 or multiple blocks.

[0090] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0091] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. An obstacle detection method based on optical flow feature fusion, characterized in that Including: Collecting continuous images during the parking process and preprocessing the collected images; Dividing the preprocessed images into multiple grids, using the optical flow tracking algorithm to track the position changes of each grid in adjacent frames, obtaining the grids containing moving objects, and fusing the grids containing moving objects to obtain the range of moving objects; According to the preset aspect ratio threshold, dividing the range of the moving object into regions, and calculating the texture of each region separately to obtain the region where the moving object is located, specifically including: Calculating and determining whether the range of the moving object conforms to the aspect ratio of a normal moving object according to the preset first aspect ratio threshold; If the range of the moving object conforms to the aspect ratio of a normal moving object, taking the range of the moving object as the region where the moving object is located; If the range of the moving object does not conform to the aspect ratio of a normal moving object, dividing the range of the moving object into k regions according to the preset second aspect ratio threshold range, where k is a natural number greater than 1; calculating the horizontal texture, vertical texture, and diagonal texture of each divided region separately, and screening out the region that most conforms to the aspect ratio of a normal moving object according to the texture calculation results of each region, and taking this region as the region where the moving object is located.

2. The method according to claim 1, wherein The preprocessing of the collected images includes: According to the current vehicle speed, thinning the continuous frame images collected during the parking process to obtain a set of continuous images with an interval of n frames, where n is a natural number greater than 1.

3. The method according to claim 1 or 2, characterized in that, The process of dividing the preprocessed images into multiple grids and using the optical flow tracking algorithm to track the position changes of each grid in adjacent frames to obtain the grids containing moving objects includes: Dividing each frame of the preprocessed image into multiple grids respectively, and obtaining the coordinates of the center point of each grid; Tracking the center point from the previous frame image to the next frame image using the optical flow tracking algorithm to obtain the corresponding tracking point; performing reverse tracking from the next frame image to the previous frame image on the tracking point using the optical flow tracking algorithm to obtain the coordinates of the corresponding reverse tracking point; Judging whether the center point coincides with its corresponding reverse tracking point according to the coordinate values: if the judgment result is coincidence, it is determined that the grid where the current center point is located is stationary; if the judgment result is non - coincidence, it is determined that the grid where the current center point is located has moved.

4. The method according to claim 3, wherein During the process of dividing the image into multiple grids, first exclude the edge area of the image according to the number of pixels, and then divide the remaining area into multiple grids.

5. The method according to claim 1, wherein The process of fusing the grids containing moving objects to obtain the range of moving objects includes: Performing dilation and erosion operations on the grids containing moving objects, then screening the grids containing moving objects again, and fusing the newly obtained grids containing moving objects to obtain the range of moving objects.

6. The method according to claim 1, characterized in that, The second aspect ratio threshold range includes multiple levels of aspect ratio threshold ranges, and each level of aspect ratio threshold range corresponds to a number of region divisions.

7. An obstacle detection system based on optical flow feature fusion, characterized in that Including: A collection and preprocessing module, used to collect continuous images during the parking process and preprocess the collected images; An optical flow tracking detection module, which is used to divide the preprocessed image into multiple grids, adopt an optical flow tracking algorithm to track the position changes of each grid in adjacent frames, obtain the grids containing moving objects, and fuse the grids containing moving objects to obtain the range of the moving object; A fine feature recognition module, which is used to divide the range of the moving object according to a preset aspect ratio threshold, calculate the texture of each region respectively to obtain the region where the moving object is located, specifically including: According to a preset first aspect ratio threshold, calculate and determine whether the range of the moving object conforms to the aspect ratio of a normal moving object; If the range of the moving object conforms to the aspect ratio of a normal moving object, use the range of the moving object as the region where the moving object is located; If the range of the moving object does not conform to the aspect ratio of a normal moving object, divide the range of the moving object into k regions according to a preset second aspect ratio threshold range, where k is a natural number greater than 1; calculate the horizontal texture, vertical texture, and oblique texture of each divided region respectively, and screen out the region that most conforms to the aspect ratio of a normal moving object according to the texture calculation results of each region, and use this region as the region where the moving object is located.

8. An electronic device, characterized in that, It includes a memory and a processor. When the processor executes a computer management program stored in the memory, it implements the steps of the obstacle detection method based on optical flow feature fusion according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, A computer management program is stored thereon. When the computer management program is executed by a processor, it implements the steps of the obstacle detection method based on optical flow feature fusion according to any one of claims 1-6.

Images