Method and system for identifying non-working areas based on images

Abstract

The present invention provides a method and system for identifying non-working areas based on an image. The method includes: obtaining an original image and separating an H-channel image and a V-channel image; converting the H-channel image into a binary image; converting the V-channel image into an edge image; based on the binary image, obtaining a minimum rectangular outline enclosing each predicted non-working area in the binary image; obtaining a first parameter representing the size of the rectangular outline, a second parameter representing the position of the rectangular outline, a third parameter representing the proportion of valid pixel values, and a fourth parameter representing the roughness of the image; and confirming whether the area in front of the robot is a true non-working area based on the relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and a preset parameter value. The present invention can accurately identify non-working areas through image recognition, save costs, and improve the performance of the robot.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control, and in particular to a method and system for identifying non-working areas based on images. Background Art

[0002] Low repetition rate and high coverage are the goals of traversal robots, such as those used for vacuuming, mowing, and pool cleaning. For example, a smart lawnmower robot operates within a lawn surrounded by a boundary, with the rest of the lawn defined as the non-operating area.

[0003] In the existing technology, the boundaries of the lawn work area are usually demarcated by burying boundary lines. This method requires a lot of manpower and material resources, which increases the cost of using mobile robots. In addition, this method has certain requirements for wiring. For example, the angle of the corner cannot be less than 90 degrees, which limits the shape of the lawn work area to a certain extent. Summary of the Invention

[0004] In order to solve the above technical problems, the object of the present invention is to provide a method and system for identifying non-working areas based on images.

[0005] In order to achieve one of the above-mentioned objects of the invention, an embodiment of the present invention provides a method for identifying a non-working area based on an image, the method comprising: acquiring an original image;

[0006] Separate the H channel image and the V channel image from the original image;

[0007] performing a binarization process on the H channel image to form a binarized image, wherein the binarized image includes at least one pre-judgment working area having a first pixel value and / or includes at least one pre-judgment non-working area having a second pixel value;

[0008] and performing edge extraction on the V channel image to form an edge image;

[0009] Based on the binary image, respectively obtaining the smallest rectangular outline enclosing each predicted non-working area in the binary image;

[0010] Acquire, based on the binary image, a first parameter representing the size of the rectangular outline, a second parameter representing the position of the rectangular outline, and a third parameter representing the proportion of effective pixel values ​​in the image;

[0011] Acquire a fourth parameter representing image roughness based on the binarized image and the edge image;

[0012] Whether the area in front of the robot is a real non-working area is determined based on the size relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value.

[0013] Through the above method, it is possible to accurately determine whether there is a non-working area in front of the robot through image recognition, saving costs and improving the robot's performance.

[0014] As a further improvement of an embodiment of the present invention, binarizing the H channel image to form a binarized image includes:

[0015] Performing filtering and normalization processing on the H channel image in sequence to obtain a first H channel preprocessed image;

[0016] performing edge extraction on the first H channel preprocessed image to form a second H channel preprocessed image;

[0017] The second H channel pre-processed image is sequentially subjected to dilation and closing operations to form a binary image.

[0018] Through the above preferred implementation manner, the process of forming a binary image is specifically described.

[0019] As a further improvement of an embodiment of the present invention, performing edge extraction on the V channel image to form an edge image includes:

[0020] Performing filtering and normalization processing on the V channel image in sequence to obtain a V channel preprocessed image;

[0021] Edge extraction is performed on the V channel preprocessed image to form an edge image.

[0022] Through the above preferred implementation manner, the process of forming the edge image is described in detail.

[0023] As a further improvement of one embodiment of the present invention, the method includes:

[0024] The first parameter A1 is configured as: at least one of the area of ​​the rectangular outline, the length of the diagonal line, the width of the rectangular outline, the height of the rectangular outline, and the number of pixels in the predicted non-working area of ​​the rectangular outline; the second parameter A2 is configured as: the coordinate parameter value of the rectangular outline; the third parameter A3 is configured as: the ratio of the number of pixels in the predicted non-working area of ​​the rectangular outline corresponding to the preset chromaticity coverage range of the H channel image to the number of pixels covered in the predicted non-working area within the rectangular outline; the fourth parameter A4 is configured as: the ratio of the number of pixels covered in the predicted non-working area of ​​the rectangular outline corresponding to the edge of the predicted non-working area of ​​the edge image to the number of pixels covered in the predicted non-working area within the rectangular outline;

[0025] Determining whether the area in front of the robot is a real non-working area based on the relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value includes:

[0026] If both:

[0027] If A1 is greater than the preset size threshold, A2 is greater than the preset coordinate threshold, A3 is less than the preset quantity threshold, and A4 is less than the preset roughness threshold, it is confirmed that the area in front of the robot is a real non-working area.

[0028] Through the above preferred implementation manner, the first parameter, the second parameter, the third parameter, and the fourth parameter are specifically defined, and the current position of the robot is accurately identified through specific rules.

[0029] As a further improvement of one embodiment of the present invention, if it is confirmed that the area in front of the robot is a real non-working area, the method further includes:

[0030] Drive the robot to execute obstacle avoidance logic.

[0031] In the above embodiment, the robot confirms that it encounters an obstacle through image recognition and performs obstacle avoidance operations, thereby improving work efficiency.

[0032] In order to achieve one of the above-mentioned objects of the invention, an embodiment of the present invention provides a system for identifying a non-working area based on an image, the system comprising: an acquisition module for acquiring an original image;

[0033] A conversion module, used for separating an H channel image and a V channel image from an original image;

[0034] performing a binarization process on the H channel image to form a binarized image, wherein the binarized image includes at least one pre-judgment working area having a first pixel value and / or includes at least one pre-judgment non-working area having a second pixel value;

[0035] and performing edge extraction on the V channel image to form an edge image;

[0036] The analysis module obtains the smallest rectangular outline enclosing each predicted non-working area in the binary image based on the binary image;

[0037] Acquire, based on the binarized image, a first parameter representing the size of the rectangular outline, a second parameter representing the position of the rectangular outline, and a third parameter representing the proportion of valid pixel values;

[0038] Acquire a fourth parameter representing image roughness based on the binarized image and the edge image;

[0039] Whether the area in front of the robot is a real non-working area is determined based on the size relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value.

[0040] Through the above system, image recognition can be used to accurately determine whether there is a non-working area in front of the robot, saving costs and improving the robot's performance.

[0041] As a further improvement of an embodiment of the present invention, the conversion module is used to: sequentially perform filtering processing and normalization processing on the H channel image to obtain a first H channel preprocessed image;

[0042] performing edge extraction on the first H channel preprocessed image to form a second H channel preprocessed image;

[0043] The second H channel pre-processed image is sequentially subjected to dilation and closing operations to form a binary image.

[0044] Through the above preferred implementation manner, the process of forming a binary image is specifically described.

[0045] As a further improvement of an embodiment of the present invention, the conversion module is used to: sequentially perform filtering processing and normalization processing on the V channel image to obtain a V channel preprocessed image;

[0046] Edge extraction is performed on the V channel preprocessed image to form an edge image.

[0047] Through the above preferred implementation manner, the process of forming the edge image is described in detail.

[0048] As a further improvement of one embodiment of the present invention, the parsing module is used to: configure the first parameter A1 to be at least one of: the area of ​​the rectangular outline, the diagonal length, the width of the rectangular outline, the height of the rectangular outline, and the number of pixels in the predicted non-working area of ​​the rectangular outline; configure the second parameter A2 to be the coordinate parameter value of the rectangular outline; configure the third parameter A3 to be the ratio of the number of pixels in the predicted non-working area of ​​the rectangular outline corresponding to the preset chromaticity coverage range of the H channel image to the number of pixels covered in the predicted non-working area within the rectangular outline; configure the fourth parameter A4 to be the ratio of the number of pixels covered in the predicted non-working area of ​​the rectangular outline corresponding to the edge of the predicted non-working area of ​​the edge image to the number of pixels covered in the predicted non-working area within the rectangular outline;

[0049] Determining whether the area in front of the robot is a real non-working area based on the relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value includes:

[0050] If both:

[0051] If A1 is greater than the preset size threshold, A2 is greater than the preset coordinate threshold, A3 is less than the preset quantity threshold, and A4 is less than the preset roughness threshold, it is confirmed that the area in front of the robot is a real non-working area.

[0052] The above system specifically defines the first parameter, the second parameter, the third parameter, and the fourth parameter, and accurately identifies the current position of the robot through specific rules.

[0053] As a further improvement of an embodiment of the present invention, if it is confirmed that the area in front of the robot is a real non-working area, the analysis module is further used to: drive the robot to execute obstacle avoidance logic.

[0054] The above system uses image recognition to confirm that the robot encounters an obstacle, performs obstacle avoidance operations, and improves work efficiency.

[0055] Compared with the prior art, the method and system for identifying non-working areas based on images of the present invention can accurately identify non-working areas through image recognition, save costs, and improve the performance of the robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 Schematic diagram of the structure of the lawn mowing robot system provided by the present invention;

[0057] Figure 2 1 is a flow chart of a method for identifying a non-working area based on an image provided by the present invention;

[0058] Figure 3 It is a module schematic diagram of the system based on image recognition of non-working areas provided by the present invention. DETAILED DESCRIPTION

[0059] The present invention will be described in detail below with reference to the various embodiments shown in the accompanying drawings. However, these embodiments do not limit the present invention, and any structural, methodological, or functional modifications made by those skilled in the art based on these embodiments are all within the scope of protection of the present invention.

[0060] The robot system of the present invention can be a lawn mowing robot system, a sweeping robot system, a snow sweeper system, a leaf vacuum system, a golf course ball picker system, etc. Each system can automatically walk in the working area and perform corresponding work. In the specific example of the present invention, the robot system is a lawn mowing robot system as an example for specific description. Accordingly, the working area can be a lawn.

[0061] like Figure 1 As shown, the lawn mowing robot system of the present invention includes: a lawn mowing robot (RM).

[0062] The lawn mower robot comprises a main body 10, a travel unit, an image acquisition unit, and a control unit disposed on the main body 10. The travel unit comprises a driving wheel 111, a driven wheel 113, and a motor for driving the driving wheel 111. The motor can be a brushless motor with a reduction gearbox and a Hall effect sensor. Once the motor is started, the reduction gearbox drives the driving wheel 111. By controlling the speed and direction of the two wheels, the robot can achieve various driving movements, including forward and backward straight-line movement, turning on the spot, and arc movement. The driven wheel 113 can be a universal wheel, typically one or two in number, and primarily serves to support and balance the vehicle.

[0063] The image acquisition unit is used to acquire scenes within its viewing angle within a certain range. In a specific embodiment of the present invention, the image acquisition unit is a camera 12. The camera 12 is installed on the upper part of the body 10 and forms a certain angle with the horizontal direction. It can capture scenes within a certain range of the lawn mower robot. The camera 12 usually captures scenes within a certain range in front of the lawn mower robot.

[0064] The control unit is a main controller 13 for image processing, such as an MCU or a DSP.

[0065] Furthermore, the lawn mowing robot also includes: a working mechanism for working, and a power supply 14; in this embodiment, the working mechanism is a lawn mowing disc, and various sensors for sensing the walking state of the walking robot, such as: tipping, lifting off the ground, collision sensors, geomagnetism, gyroscopes, etc., which are not described in detail here.

[0066] like Figure 2 As shown, the first embodiment of the present invention provides a method for identifying a non-working area based on an image, the method comprising the following steps:

[0067] S1, obtain the original image;

[0068] S2, separating the H channel image and the V channel image from the original image;

[0069] performing a binarization process on the H channel image to form a binarized image, wherein the binarized image includes at least one pre-judgment working area having a first pixel value and / or includes at least one pre-judgment non-working area having a second pixel value;

[0070] and performing edge extraction on the V channel image to form an edge image;

[0071] S3. Based on the binary image, obtaining the smallest rectangular outline enclosing each predicted non-working area in the binary image;

[0072] Acquire, based on the binary image, a first parameter representing the size of the rectangular outline, a second parameter representing the position of the rectangular outline, and a third parameter representing the proportion of effective pixel values ​​in the image;

[0073] Acquire a fourth parameter representing image roughness based on the binarized image and the edge image;

[0074] Whether the area in front of the robot is a real non-working area is determined based on the size relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value.

[0075] Whether the area in front of the robot is a real non-working area is determined based on the size relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value.

[0076] In a specific embodiment of the present invention, in step S1, a camera mounted on the lawn mowing robot captures the scene in front of the robot in real time, generating a raw image. The scene is an image of the ground in the robot's forward direction. Furthermore, upon receiving the raw image, the main controller analyzes the raw image to determine the robot's working position at the time the raw image was captured. This is described in detail below. In this specific example, the raw image format is not specifically limited, and may be, for example, a color image in RGB or HSV format.

[0077] For step S2, if the original image is in RGB format, the original image in RGB format is format converted to form an HSV image. If the original image is in HSV format, no conversion is required, and the H channel image and the V channel image are directly separated from the HSV image. This implementation method is all existing technology, and there are many implementation methods, which will not be described here.

[0078] The H channel image is binarized to form a binary image. It can be seen that the H channel image is a grayscale image. The converted binary image is: the pixels in the image have only two grayscale values, for example, the two grayscale values ​​are 0 and 255 respectively. The binarization process is a process of presenting the entire original image with a clear black and white effect.

[0079] In a preferred embodiment of the present invention, binarizing the H-channel image to form a binary image includes: sequentially filtering and normalizing the H-channel image to obtain a first H-channel preprocessed image to remove noise in the H-channel image, performing edge extraction on the first H-channel preprocessed image to form a second H-channel preprocessed image; for example, the edge detection method used is the Canny algorithm; and sequentially performing dilation and closing operations on the second H-channel preprocessed image to form a binary image; the dilation and closing operations are both conventional image processing methods, which are also existing technologies and will not be further described herein.

[0080] Preferably, performing edge extraction on the V-channel image to form an edge image includes: performing filtering processing and normalization processing on the V-channel image in sequence to obtain a V-channel preprocessed image to remove noise in the V-channel image, and performing edge extraction on the V-channel preprocessed image to form an edge image. For example, the edge detection method used is the Canny algorithm.

[0081] Preferably, for step S3, the method includes: configuring the first parameter A1 to be at least one of: the area of ​​the rectangular outline, the length of the diagonal line, the width of the rectangular outline, the height of the rectangular outline, and the number of pixels in the predicted non-working area of ​​the rectangular outline; configuring the second parameter A2 to be the coordinate parameter value of the rectangular outline; configuring the third parameter A3 to be the ratio of the number of pixels in the predicted non-working area of ​​the rectangular outline corresponding to the preset chromaticity coverage range of the H channel image to the number of pixels covered in the predicted non-working area within the rectangular outline; configuring the fourth parameter A4 to be the ratio of the number of pixels covered in the predicted non-working area of ​​the rectangular outline corresponding to the edge of the predicted non-working area of ​​the edge image to the number of pixels covered in the predicted non-working area within the rectangular outline;

[0082] Determining whether the area in front of the robot is a real non-working area based on the relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value includes:

[0083] If both:

[0084] If A1 is greater than the preset size threshold, A2 is greater than the preset coordinate threshold, A3 is less than the preset quantity threshold, and A4 is less than the preset roughness threshold, it is confirmed that the area in front of the robot is a real non-working area.

[0085] Here, when the first parameter A1 is the area corresponding to the rectangular outline, the preset size threshold is also the corresponding area threshold; correspondingly, when the first parameter A1 is at least one of the diagonal length, the rectangular outline width, the rectangular outline height, and the number of pixels in the rectangular outline that are predicted to be in the non-working area, the first parameter A1 also changes accordingly. Specifically, the preset size threshold is one of the diagonal length threshold, the rectangular outline width threshold, the rectangular outline height threshold, and the number of pixels in the rectangular outline that are predicted to be in the non-working area;

[0086] In a specific example of the present invention, the configuration area threshold is 5%*the area of ​​the original image, the diagonal length threshold is 22%*the diagonal length of the original image, the pixel number threshold for predicting the non-working area in the rectangular outline is 5%*the number of pixels in the original image; the rectangular outline width threshold is 50%*the width of the original image, and the rectangular outline height is 50% of the height of the original image.

[0087] The second parameter A2 is configured as the coordinate parameter value of the rectangular outline. The coordinate parameter value can be the y-axis coordinate of the lower right corner of the rectangular outline, or the center position coordinate of the rectangular outline, or the average value of the Y value of the pixel coordinate in the first outline, etc. No further details are given here. Accordingly, in a specific example of the present invention, the upper left corner is the coordinate origin, the Y axis is downward as the positive direction of the Y axis, and the preset coordinate threshold is 62.6%*the coordinate value of the height of the original image.

[0088] The third parameter A3 is configured as the ratio of valid pixel values ​​in the rectangular outline, i.e., the ratio of the number of pixels in the predicted non-working area of ​​the rectangular outline corresponding to the preset chromaticity coverage range of the H-channel image to the number of pixels covered by the predicted non-working area within the rectangular outline. The preset chromaticity coverage range of the H-channel image is generally the chromaticity value range of the common color of lawn, such as the green chromaticity range. Preferably, the value of A3 is 0.8.

[0089] The fourth parameter A4 is configured as the roughness of the rectangular outline, i.e., the ratio of the number of pixels covered by the predicted non-working area of ​​the rectangular outline corresponding to the edge of the predicted non-working area of ​​the edge image to the number of pixels covered by the predicted non-working area within the rectangular outline. Preferably, the value of A4 is 0.26.

[0090] Furthermore, if it is confirmed that the area in front of the robot is a real non-working area, the method further includes: driving the robot to execute obstacle avoidance logic, such as backward steering, which will not be described in detail here.

[0091] Combine Figure 3 As shown, a system for identifying non-working areas based on images is provided, and the system includes: an acquisition module 100, a conversion module 200 and a parsing module 300.

[0092] The acquisition module 100 is used to acquire the original image;

[0093] The conversion module 200 is used to separate the H channel image and the V channel image from the original image;

[0094] performing a binarization process on the H channel image to form a binarized image, wherein the binarized image includes at least one pre-judgment working area having a first pixel value and / or includes at least one pre-judgment non-working area having a second pixel value;

[0095] and performing edge extraction on the V channel image to form an edge image;

[0096] The analysis module 300 is used to obtain the smallest rectangular outline enclosing each predicted non-working area in the binary image based on the binary image;

[0097] Acquire, based on the binarized image, a first parameter representing the size of the rectangular outline, a second parameter representing the position of the rectangular outline, and a third parameter representing the proportion of valid pixel values;

[0098] Acquire a fourth parameter representing image roughness based on the binarized image and the edge image;

[0099] Whether the area in front of the robot is a real non-working area is determined based on the size relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value.

[0100] Specifically, the acquisition module 100 is used to implement step S1; the conversion module 200 is used to implement step S2; the analysis module 300 is used to implement step S3, and to execute the calculation of the obstacle avoidance logic for driving the robot; technical personnel in the relevant field can clearly understand that for the convenience and conciseness of description, the specific working process of the system described above can refer to the corresponding process in the aforementioned method implementation method, and will not be repeated here.

[0101] In summary, the method and system for identifying non-working areas based on images of the present invention can accurately identify non-working areas through image recognition, save costs, and improve the performance of the robot.

[0102] In the several embodiments provided in this application, it should be understood that the disclosed modules, systems, and methods can be implemented in other ways. The system implementation described above is merely illustrative, and the module division is merely a logical functional division. In actual implementation, other division methods may be used, for example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not implemented.

[0103] The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of this embodiment.

[0104] In addition, the functional modules in various embodiments of the present application may be integrated into a single parsing module, or each module may exist physically separately, or two or more modules may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A method for identifying non-working areas based on images, characterized in that: The method comprises: Get the original image; Separate the H channel image and the V channel image from the original image; performing a binarization process on the H channel image to form a binarized image, wherein the binarized image includes at least one pre-judgment working area having a first pixel value and / or includes at least one pre-judgment non-working area having a second pixel value; and performing edge extraction on the V channel image to form an edge image; Based on the binary image, respectively obtaining the smallest rectangular outline enclosing each predicted non-working area in the binary image; Acquire, based on the binary image, a first parameter representing the size of the rectangular outline, a second parameter representing the position of the rectangular outline, and a third parameter representing the proportion of effective pixel values ​​in the image; Acquire a fourth parameter representing image roughness based on the binarized image and the edge image; Determine whether the area in front of the robot is a real non-working area based on the relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value; The method comprises: The first parameter A1 is configured as: at least one of the area of ​​the rectangular outline, the length of the diagonal line, the width of the rectangular outline, the height of the rectangular outline, and the number of pixels in the predicted non-working area of ​​the rectangular outline; the second parameter A2 is configured as: the coordinate parameter value of the rectangular outline; the third parameter A3 is configured as: the ratio of the number of pixels in the predicted non-working area of ​​the rectangular outline corresponding to the preset chromaticity coverage range of the H channel image to the number of pixels covered in the predicted non-working area within the rectangular outline; the fourth parameter A4 is configured as: the ratio of the number of pixels covered in the predicted non-working area of ​​the rectangular outline corresponding to the edge of the predicted non-working area of ​​the edge image to the number of pixels covered in the predicted non-working area within the rectangular outline; Determining whether the area in front of the robot is a real non-working area based on the relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value includes: If both: If A1 is greater than the preset size threshold, A2 is greater than the preset coordinate threshold, A3 is less than the preset quantity threshold, and A4 is less than the preset roughness threshold, it is confirmed that the area in front of the robot is a real non-working area.

2. The method for identifying non-working areas based on images according to claim 1, characterized in that: Binarizing the H channel image to form a binary image includes: Performing filtering and normalization processing on the H channel image in sequence to obtain a first H channel preprocessed image; performing edge extraction on the first H channel preprocessed image to form a second H channel preprocessed image; The second H channel pre-processed image is sequentially subjected to dilation and closing operations to form a binary image.

3. The method for identifying non-working areas based on images according to claim 1, characterized in that: Performing edge extraction on the V channel image to form an edge image includes: Performing filtering and normalization processing on the V channel image in sequence to obtain a V channel preprocessed image; Edge extraction is performed on the V channel preprocessed image to form an edge image.

4. The method for identifying a non-working area based on an image according to any one of claims 1 to 3, characterized in that: If it is confirmed that the area in front of the robot is a real non-working area, the method further includes: Drive the robot to execute obstacle avoidance logic.

5. A system for identifying non-working areas based on images, characterized in that: The system comprises: An acquisition module, used to acquire the original image; A conversion module, used for separating an H channel image and a V channel image from an original image; performing a binarization process on the H channel image to form a binarized image, wherein the binarized image includes at least one pre-judgment working area having a first pixel value and / or includes at least one pre-judgment non-working area having a second pixel value; and performing edge extraction on the V channel image to form an edge image; The analysis module obtains the smallest rectangular outline enclosing each predicted non-working area in the binary image based on the binary image; Acquire, based on the binarized image, a first parameter representing the size of the rectangular outline, a second parameter representing the position of the rectangular outline, and a third parameter representing the proportion of valid pixel values; Acquire a fourth parameter representing image roughness based on the binarized image and the edge image; Determine whether the area in front of the robot is a real non-working area based on the relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value; The parsing module is used to: configure the first parameter A1 to be at least one of: the area of ​​the rectangular outline, the length of the diagonal line, the width of the rectangular outline, the height of the rectangular outline, and the number of pixels in the predicted non-working area of ​​the rectangular outline; configure the second parameter A2 to be the coordinate parameter value of the rectangular outline; configure the third parameter A3 to be the ratio of the number of pixels in the predicted non-working area of ​​the rectangular outline corresponding to the preset chromaticity coverage range of the H channel image to the number of pixels covered in the predicted non-working area within the rectangular outline; configure the fourth parameter A4 to be the ratio of the number of pixels covered in the predicted non-working area of ​​the rectangular outline corresponding to the edge of the predicted non-working area of ​​the edge image to the number of pixels covered in the predicted non-working area within the rectangular outline; Determining whether the area in front of the robot is a real non-working area based on the relationship between the first parameter, the second parameter, the third parameter, the fourth parameter and the preset parameter value includes: If both: If A1 is greater than the preset size threshold, A2 is greater than the preset coordinate threshold, A3 is less than the preset quantity threshold, and A4 is less than the preset roughness threshold, it is confirmed that the area in front of the robot is a real non-working area.

6. The system for identifying non-working areas based on images according to claim 5, characterized in that: The conversion module is used to: Performing filtering and normalization processing on the H channel image in sequence to obtain a first H channel preprocessed image; performing edge extraction on the first H channel preprocessed image to form a second H channel preprocessed image; The second H channel pre-processed image is sequentially subjected to dilation and closing operations to form a binary image.

7. The system for identifying non-working areas based on images according to claim 5, characterized in that: The conversion module is used to: perform filtering and normalization processing on the V channel image in sequence to obtain a V channel preprocessed image; Edge extraction is performed on the V channel preprocessed image to form an edge image.

8. The system for identifying non-working areas based on images according to any one of claims 5 to 7, characterized in that: If it is confirmed that the area in front of the robot is a real non-working area, the parsing module is also used to drive the robot to execute obstacle avoidance logic.

Images