Method, system and robot based on an image recognition working area

Abstract

The present invention provides a method, a system and a robot for identifying a working area based on image recognition. The method includes: obtaining an original image; separating an H-channel image from the original image, statistically analyzing the histogram of the H-channel image, and obtaining a first parameter representing the peak value of the histogram; performing binarization processing on the original image to form a first binarized image; the first binarized image includes a working area and a non-working area with different pixel values; statistically analyzing a second parameter representing the size of the non-working area in the first binarized image based on the first binarized image; and identifying the working area according to the magnitude relationship between the first parameter, the second parameter and a preset parameter threshold. The present invention can accurately identify the working area through image recognition, save costs and improve the usage performance of the robot.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control, and particularly to a method, a system and a robot for identifying a working area based on image recognition. Background Art

[0002] Low repetition rate and high coverage rate are the goals pursued by traversing robots such as mobile robots for vacuuming, lawn mowing and pool cleaning. Taking a mobile robot as an intelligent lawn mowing robot as an example, the lawn mowing robot uses the lawn surrounded by a boundary as a working area for mowing operations, and the area outside the lawn is defined as a non-working area.

[0003] In the prior art, the boundary of the lawn working area is usually calibrated by burying a boundary line. This method requires a lot of manpower and material resources, increasing the use cost of the mobile robot; moreover, this method has certain requirements for wiring. For example, the angle of the corner cannot be less than 90 degrees, so to a certain extent, the shape of the lawn working area is restricted. Summary of the Invention

[0004] To solve the above technical problems, the object of the present invention is to provide a method, a system and a robot for identifying a working area based on image recognition.

[0005] To achieve one of the above invention objects, an embodiment of the present invention provides a method for identifying a working area based on image recognition, the method comprising: acquiring an original image;

[0006] Separating an H-channel image from the original image;

[0007] Performing binarization processing on the original image to form a first binarized image; the first binarized image includes a working area and a non-working area with different pixel values;

[0008] Statistical histogram of the H-channel image, and obtaining a first parameter representing the peak value of the histogram;

[0009] Based on the first binarized image, statistically obtaining a second parameter representing the size of the non-working area in the first binarized image;

[0010] Identifying the working area according to the magnitude relationship between the first parameter, the second parameter and a preset parameter threshold;

[0011] By the above method, the current position of the robot can be accurately judged through image recognition, saving costs and improving the use performance of the robot.

[0012] As a further improvement of an embodiment of the present invention, the first parameter is: the histogram peak value maxH of the H-channel image; the second parameter is: the number N1 of non-working area pixel points in the first binary image; the preset parameter threshold includes: a preset first value M1 and a preset first peak value H1, and / or includes a preset second value M2 and a preset second peak value H2;

[0013] Identifying the working area according to the magnitude relationship between the first parameter, the second parameter and the preset parameter value, includes:

[0014] If both are satisfied: maxH > H1, N1 > M1, then it is determined as a non-working area;

[0015] Or if both are satisfied: maxH ≤ H2, N1 > M2, then it is determined as a non-working area;

[0016] Wherein, H1 = H2, M1 < M2;

[0017] Through the above preferred embodiment, the first parameter and the second parameter are specifically defined, and the current position of the robot is accurately identified through specific rules.

[0018] As a further improvement of an embodiment of the present invention, the method further includes:

[0019] Inverting the first binary image to form a second binary image; the second binary image includes a working area and a non-working area with different pixel values;

[0020] Based on the second binary image, obtain the smallest rectangular contour enclosing the non-working area of the second binary image;

[0021] Based on the second binary image, obtain the rectangular contour coordinate parameter value Y;

[0022] If it is confirmed that the robot is currently in a non-working area and Y is greater than the preset coordinate value, then drive the robot to execute the obstacle avoidance logic;

[0023] Wherein, the value range of the preset coordinate value is [58%*(L2*W2), 66%*(L2*W2)], and L2 and W2 respectively represent the length and width of the second binary image;

[0024] Based on the identification of the current position of the robot, it is judged whether the robot encounters an obstacle through image recognition, so as to improve the working efficiency.

[0025] To achieve one of the above invention purposes, an embodiment of the present invention provides a method for identifying a working area based on an image, the method includes:

[0026] Obtain the original image;

[0027] Separate the V-channel image from the original image, perform edge extraction on the V-channel image to form an edge image, and separate the H-channel image from the HSV image, and statistically analyze the histogram of the H-channel image;

[0028] Perform binarization processing on the original image to form a first binarized image, and invert the first binarized image to form a second binarized image; both the first binarized image and the second binarized image include a working area and a non-working area with different pixel values;

[0029] Based on the second binarized image, obtain the smallest rectangular contour that encloses the non-working area of the second binarized image;

[0030] Obtain a first parameter representing the peak value of the histogram of the H-channel image based on the H-channel image;

[0031] Statistically analyze a third parameter representing the roughness of the edge image within the range of the non-working area enclosed by the rectangular contour corresponding to the edge image;

[0032] Obtain a fourth parameter representing the attributes of the rectangular contour based on the second binarized image;

[0033] Identify the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter values;

[0034] Through the above method, the current position of the robot can be accurately judged by image recognition, saving costs and improving the performance of the robot.

[0035] As a further improvement of an embodiment of the present invention, the first parameter is: the peak value maxH of the histogram of the H-channel image; the third parameter is the ratio P1 of the number of edge pixel points in the edge image corresponding to the non-working area within the rectangular contour to the number of pixel points in the non-working area within the rectangular contour; the fourth parameter includes: the size DL of the rectangular contour, the number N3 of pixel points in the non-working area within the rectangular contour, and the ratio P2 of the number of pixel points in the working area within the rectangular contour to the number of pixel points in the non-working area; the size of the rectangular contour includes at least one of: the diagonal length XL of the rectangular contour, the length LL of the rectangular contour, and the width WL of the rectangular contour;

[0036] Configuring the preset parameter thresholds includes: a preset third peak value H3, a preset fourth peak value H4, a preset first ratio Q1, a preset first length value Le1, a preset third quantity M3, and a preset second ratio Q2;

[0037] And / or includes; a preset third peak value H3, a preset fourth peak value H4, a preset first ratio Q1, a preset first length value Le1, a preset fourth quantity M4, and a preset third ratio Q3;

[0038] and / or include; preset third peak value H3, preset fourth peak value H4, preset first ratio Q1, preset first length value Le1, preset fifth quantity M5, preset fourth ratio Q4;

[0039] and / or include; preset third peak value H3, preset fourth peak value H4, preset first ratio Q1, preset first length value Le1, preset sixth quantity M6, preset fifth ratio Q5;

[0040] and / or include; preset third peak value H3, preset fourth peak value H4, preset sixth ratio Q6, preset first length value Le1, preset sixth quantity M6, preset seventh ratio Q7;

[0041] Identify the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter value, including:

[0042] If all of the following conditions are met simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M3, P2 < Q2, then it is determined as a non-working area;

[0043] Or if all of the following conditions are met simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M4, P2 < Q3, then it is determined as a non-working area;

[0044] Or if all of the following conditions are met simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M5, P2 < Q4, then it is determined as a non-working area;

[0045] Or if all of the following conditions are met simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M6, P2 < Q5, then it is determined as a non-working area;

[0046] Or if all of the following conditions are met simultaneously: H3 < maxH < H4, P1 < Q6, DL > Le1, N3 > M6, P2 < Q7, then it is determined as a non-working area;

[0047] Wherein, M3 < M4 < M5 < M6, Q2 < Q3 < Q4 < Q5 < Q7, Q1 > Q6;

[0048] Through the above preferred embodiments, the first parameter, the third parameter and the fourth parameter are specifically defined, and the current position of the robot is accurately identified through specific rules.

[0049] As a further improvement of an embodiment of the present invention, the first parameter is: the histogram peak maxH of the H-channel image; the third parameter is the ratio P1 of the number of edge pixel points in the edge image corresponding to the non-working area within the rectangular contour to the number of pixel points in the non-working area within the rectangular contour; the fourth parameter includes: the rectangular contour size DL, the number N3 of pixel points in the non-working area within the rectangular contour, the ratio P2 of the number of pixel points in the working area within the rectangular contour to the number of pixel points in the non-working area, and the ratio P3 of the number of pixel points in the non-working area within the rectangular contour corresponding to the working area in the first binary image to the number of pixel points in the non-working area within the rectangular contour; the rectangular contour size includes at least one of the diagonal length XL of the rectangular contour, the length LL of the rectangular contour, and the width WL of the rectangular contour.

[0050] The preset parameter thresholds include: a preset fourth peak H4, a preset fifth peak H5, a preset eighth ratio Q8, a preset first length value Le1, a preset seventh quantity M7, a preset ninth ratio Q9, and a preset tenth ratio Q10.

[0051] Identifying the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter values includes:

[0052] If all of the following conditions are satisfied simultaneously: H4 < maxH < H5, P1 < Q8, DL > Le1, N3 > M7, P2 < Q9, P3 < Q10, then it is determined as a non-working area.

[0053] Through the above preferred embodiment, the first parameter, the third parameter and the fourth parameter are specifically defined, and the current position of the robot is accurately identified through specific rules.

[0054] As a further improvement of an embodiment of the present invention, the first parameter is: the histogram peak maxH of the H-channel image; the third parameter is the ratio P1 of the number of edge pixel points in the edge image corresponding to the non-working area within the rectangular contour to the number of pixel points in the non-working area within the rectangular contour; the fourth parameter includes: the rectangular contour size DL, the ratio P2 of the number of pixel points in the working area within the rectangular contour to the number of pixel points in the non-working area; the ratio P3 of the number of pixel points in the non-working area within the rectangular contour corresponding to the working area in the first binary image to the number of pixel points in the non-working area within the rectangular contour; the rectangular contour size includes at least one of the diagonal length XL of the rectangular contour, the length LL of the rectangular contour, and the width WL of the rectangular contour.

[0055] The preset parameter thresholds include: a preset fifth peak value H5, a preset eleventh ratio Q11, a preset first length value Le1, a preset twelfth ratio Q12, and a preset thirteenth ratio Q13;

[0056] and / or a preset fifth peak value H5, a preset fourteenth ratio Q14, a preset first length value Le1, a preset fifteenth ratio Q15, and a preset thirteenth ratio Q13;

[0057] and / or a preset fifth peak value H5, a preset sixteenth ratio Q16, a preset first length value Le1, a preset seventeenth ratio Q17, and a preset thirteenth ratio Q13;

[0058] Identifying the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter values includes:

[0059] If all of the following conditions are met simultaneously: maxH≥H5, P1<Q11, DL>Le1, P2<Q12, P3<Q13, then it is determined as a non-working area;

[0060] Or if all of the following conditions are met simultaneously: maxH≥H5, P1<Q14, DL>Le1, P2<Q15, P3≥Q13, then it is determined as a non-working area;

[0061] If all of the following conditions are met simultaneously: maxH≥H5, P1<Q16, DL>Le1, P2<Q17, P3≥Q13, then it is determined as a non-working area;

[0062] Among them, Q11>Q14>Q16, and Q12<Q15<Q17;

[0063] Through the above preferred embodiments, the first parameter, the third parameter and the fourth parameter are specifically defined, and the current position of the robot is accurately identified through specific rules.

[0064] As a further improvement of an embodiment of the present invention, the first parameter is: the histogram peak value maxH of the H-channel image; the third parameter is the ratio P1 of the number of edge pixel points corresponding to the non-working area within the rectangular contour in the edge image to the number of pixel points in the non-working area within the rectangular contour; the fourth parameter includes: the width WL of the rectangular contour, the ratio P2 of the number of pixel points in the working area to the number of pixel points in the non-working area within the rectangular contour; the ratio P3 of the number of pixel points corresponding to the non-working area within the rectangular contour in the first binary image on the working area to the number of pixel points in the non-working area within the rectangular contour;

[0065] The preset parameter thresholds include: a preset fifth peak value H5, a preset eleventh ratio Q11, a preset third length value Le3, a preset twelfth ratio Q12, and a preset thirteenth ratio Q13;

[0066] and / or a preset fifth peak value H5, a preset fourteenth proportion Q14, a preset third length value Le3, a preset fifteenth proportion Q15, a preset thirteenth proportion Q13;

[0067] and / or a preset fifth peak value H5, a preset sixteenth proportion Q16, a preset third length value Le3, a preset seventeenth proportion Q17, a preset thirteenth proportion Q13;

[0068] Identify the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter values, including:

[0069] If all of the following conditions are met simultaneously: maxH≥H5, P1<Q11, WL>Le3, P2<Q12, P3<Q13, then it is determined as a non-working area;

[0070] Or if all of the following conditions are met simultaneously: maxH≥H5, P1<Q14, WL>Le3, P2<Q15, P3≥Q13, then it is determined as a non-working area;

[0071] If all of the following conditions are met simultaneously: maxH≥H5, P1<Q16, WL>Le3, P2<Q17, P3≥Q13, then it is determined as a non-working area;

[0072] wherein, Q11>Q14>Q16, Q12<Q15<Q17;

[0073] Through the above preferred embodiments, the first parameter, the third parameter and the fourth parameter are specifically defined, and the current position of the robot is accurately identified through specific rules.

[0074] To achieve one of the above invention purposes, an embodiment of the present invention provides a system for identifying a working area based on image recognition, and the system includes:

[0075] An acquisition module, configured to acquire an original image;

[0076] A conversion module, configured to separate an H-channel image from the original image;

[0077] Perform binarization processing on the original image to form a first binarized image; the first binarized image includes a working area and a non-working area with different pixel values;

[0078] An analysis module, configured to count the histogram of the H-channel image and obtain a first parameter representing the peak value of the histogram;

[0079] Based on the first binarized image, count a second parameter representing the size of the non-working area in the first binarized image;

[0080] Identify the working area according to the magnitude relationship between the first parameter, the second parameter, and the preset parameter threshold;

[0081] Through the above system, the current position of the robot can be accurately judged by image recognition, saving costs and improving the performance of the robot.

[0082] To achieve one of the above invention purposes, an embodiment of the present invention provides a system for identifying a working area based on image recognition, the system includes:

[0083] An acquisition module, configured to acquire an original image;

[0084] A conversion module, configured to separate a V-channel image from the original image, perform edge extraction on the V-channel image to form an edge image, and separate an H-channel image from the HSV image, and count the histogram of the H-channel image;

[0085] Perform binarization processing on the original image to form a first binarized image, and invert the first binarized image to form a second binarized image; both the first binarized image and the second binarized image include a working area and a non-working area with different pixel values;

[0086] Based on the second binarized image, obtain the smallest rectangular contour that encloses the non-working area of the second binarized image;

[0087] An analysis module, configured to obtain a first parameter representing the peak value of the histogram of the H-channel image based on the H-channel image;

[0088] Statistically obtain a third parameter representing the roughness of the edge image within the range of the non-working area enclosed by the rectangular contour corresponding to the edge image;

[0089] Obtain a fourth parameter representing the attributes of the rectangular contour based on the second binarized image;

[0090] Identify the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter, and the preset parameter value;

[0091] Through the above system, the current position of the robot can be accurately judged by image recognition, saving costs and improving the performance of the robot.

[0092] To achieve one of the above invention purposes, an embodiment of the present invention provides a robot, including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the method for identifying a working area based on image recognition as described above are implemented.

[0093] Compared with the prior art, the method, system and robot for working area based on image recognition of the present invention can accurately recognize the working area through image recognition, save costs and improve the performance of the robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] Figure 1 is a schematic structural diagram of the lawn mowing robot system provided by the present invention;

[0095] Figure 2 is a schematic flowchart of the method for working area based on image recognition provided by the first embodiment of the present invention;

[0096] Figure 3 is a schematic flowchart of the method for working area based on image recognition provided by the second embodiment of the present invention;

[0097] Figure 4 is a schematic module diagram of the system for working area based on image recognition provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0098] The present invention will be described in detail below with reference to the embodiments shown in the drawings. However, these embodiments do not limit the present invention, and any structural, method, or functional transformation made by those of ordinary skill in the art based on these embodiments is included in the protection scope of the present invention.

[0099] The robot system of the present invention can be a lawn mowing robot system, a floor sweeping robot system, a snow sweeper system, a leaf blower system, a golf course ball picker system, etc. Each system can automatically walk in the working area and perform corresponding work. In a specific example of the present invention, the robot system is taken as a lawn mowing robot system for specific illustration. Correspondingly, the working area can be a lawn.

[0100] As Figure 1 shown, the lawn mowing robot system of the present invention includes: a lawn mowing robot (RM).

[0101] The lawn mowing robot includes: a main body 10, a walking unit, an image acquisition unit and a control unit arranged on the main body 10. The walking unit includes: 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 sensor; after the motor is started, it can drive the driving wheel 111 to walk through the reduction gearbox, and by controlling the speed and direction of the two wheels, forward and backward linear running, in-situ turning and circular running and other driving actions can be realized; the driven wheel 113 can be a universal wheel, which is usually set to 1 or 2, and its main function is to support and balance.

[0102] The image acquisition unit is used to acquire the scene within its viewing range within a certain range. In the specific embodiment of the present invention, it is the camera 12, which is installed on the upper part of the body 10 and forms a certain angle with the horizontal direction, and can capture the scene within a certain range of the lawn mowing robot; the camera 12 usually captures the scene within a certain range in front of the lawn mowing robot.

[0103] The control unit is the main controller 13 for image processing, such as: MCU or DSP, etc.

[0104] Furthermore, the lawn mowing robot further includes: a working mechanism for working, and a power supply 14; in this embodiment, the working mechanism is a mowing cutter head, and various sensors for sensing the walking state of the walking robot, such as: tilt, ground clearance, collision sensors, geomagnetism, gyroscope, etc., are not specifically described one by one here.

[0105] As Figure 2 shown, the method for working position based on image recognition provided by the first embodiment of the present invention includes the following steps:

[0106] S1. Acquire the original image;

[0107] S2. Separate the H-channel image from the original image;

[0108] Perform binary processing on the original image to form a first binary image; the first binary image includes a working area and a non-working area with different pixel values;

[0109] S3. Statistically analyze the histogram of the H-channel image and obtain a first parameter representing the peak value of the histogram;

[0110] Statistically analyze a second parameter representing the size of the non-working area in the first binary image based on the first binary image.

[0111] S4. Identify the working area according to the magnitude relationship between the first parameter, the second parameter and the preset parameter threshold.

[0112] In the specific embodiment of the present invention, for step S1, the camera installed on the lawn mowing robot captures the scene in front of the robot in real time to form an original image; the scene is the ground image in the forward direction of the robot; furthermore, when the main controller receives the original image, it parses the original image; thus, the working position of the robot for capturing the original image can be judged through the original image, which will be described in detail below. In this specific example, the format of the original image is not specifically limited, for example, it is a color image in RGB format or HSV format.

[0113] For step S2, if the original image is in RGB format, the original RGB image is converted to an HSV image. If the original image is in HSV format, no conversion is required and the H channel image is directly separated from the HSV image. These implementation methods are all prior arts and there are various ways to implement them, which will not be elaborated here.

[0114] The original image is binarized to form a first binary image. It is known that: the binary image is an image in which pixel points have only two gray values. For example, the two gray values are 0 and 255 respectively. The binarization process is a process in which the entire original image presents an obvious black and white effect.

[0115] There are many ways to implement the binarization of the original image to form a first binary image. In the specific implementation of the present invention, the original image can be segmented to obtain a chromaticity segmentation threshold [lowValue, highValue]. Pixel points within this segmentation threshold in the original image are converted to the same pixel, and pixel points outside the segmentation threshold are converted to another pixel to form a binary image. In this binary image, one pixel value represents the working area and the other pixel value represents the non - working area.

[0116] The image segmentation methods include, for example, dynamic segmentation according to color, segmentation according to edge texture method, fixed threshold method segmentation, Otsu threshold method segmentation, etc.

[0117] The way of threshold matching is to transform each pixel point by setting a threshold range. In the specific example of the present invention, the pixel values of pixel points within the threshold range are adjusted to 0, that is, the working area is adjusted to black; the pixel values of pixel points outside the threshold range are adjusted to 255, that is, the non - working area is adjusted to white.

[0118] For step S3, when the H channel image is known, the method of statistically analyzing the histogram of the H channel image is a prior art and there are various ways to implement it, which will not be elaborated here.

[0119] In a preferred embodiment of the present invention, based on the histogram, a first parameter representing the peak value of the histogram is obtained. There are various specific forms of the first parameter. In the specific implementation of the present invention, the first parameter is the peak value maxH of the histogram of the H channel image.

[0120] Correspondingly, there are also various specific forms of a second parameter representing the size of the non - working area in the first binary image. In the specific implementation of the present invention, the second parameter is the number N1 of pixel points in the non - working area of the first binary image.

[0121] It should be noted that for the above steps S2 and S3, they can be executed successively in the order of steps S2 and S3, or can be executed alternately. For example, after obtaining the H-channel image, the first parameter representing the peak value of the histogram is obtained immediately. The order of execution of steps S2 and S3 does not affect the final output result.

[0122] For step S4, the preset parameter thresholds include: a preset first value M1 and a preset first peak value H1, and / or include a preset second value M2 and a preset second peak value H2;

[0123] Identifying the working area according to the magnitude relationship between the first parameter, the second parameter and the preset parameter value includes:

[0124] a. If both of the following conditions are met: maxH > H1 and N1 > M1, it is determined as a non-working area;

[0125] Or b. If both of the following conditions are met: maxH ≤ H2 and N1 > M2, it is determined as a non-working area; where H1 < H2 and M1 < M2.

[0126] Preferably, the method includes: configuring H1 = 2.08% * (L1 * W1), N1 = 28.6% * (L1 * W1); and / or configuring H2 = 2.08% * (L1 * W1), N2 = 41.6% * (L1 * W1); where L1 and W1 respectively represent the length and width of the first binary image.

[0127] In this first embodiment, the larger the value of N1, the larger the non-working area existing in the image, the more dispersed the color distribution, and the less likely the concentration degree of a certain color.

[0128] The method for identifying the working area based on an image according to the first embodiment of the present invention can directly identify the working area based on the peak value obtained from the H-channel histogram and the comparison between the number of non-working area pixel points N1 obtained from the first binary image and the preset parameters, with a small amount of calculation and high calculation accuracy, improving the recognition and working efficiency of the robot.

[0129] Combined Figure 3 As shown, the method for identifying the working area based on an image provided by the second embodiment of the present invention includes: M1. Obtaining the original image;

[0130] M2. Separate the V-channel image from the original image, perform edge extraction on the V-channel image to form an edge image, and separate the H-channel image from the HSV image, and statistically analyze the histogram of the H-channel image; perform binarization processing on the original image to form a first binarized image, and invert the first binarized image to form a second binarized image; both the first binarized image and the second binarized image include a working area and a non-working area with different pixel values; based on the second binarized image, obtain the smallest rectangular contour that encloses the non-working area of the second binarized image.

[0131] M3. Obtain a first parameter representing the peak value of the histogram of the H-channel image based on the H-channel image; statistically analyze a third parameter representing the roughness of the edge image within the range of the non-working area enclosed by the rectangular contour corresponding to the edge image; obtain a fourth parameter representing the attributes of the rectangular contour based on the second binarized image.

[0132] M4. Identify the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter value.

[0133] For step M2, the format of the original image is the same as that of the first embodiment above. Separating the v-channel image from the HSV image and separating the H-channel image from the HSV image and statistically analyzing the histogram of the H-channel image are all prior arts, and there are various implementation methods, which will not be elaborated here.

[0134] 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 to remove noise in the V-channel image, and then extracting the edges of the denoised image through an edge detection algorithm to form an edge image; preferably, the edge detection method used is the canny algorithm.

[0135] The implementation method of performing binarization processing on the original image to form a first binarized image is the same as that of the first embodiment, and will not be further elaborated here.

[0136] Preferably, inverting the first binarized image to form a second binarized image includes: sequentially performing an inversion operation, an opening operation, and a closing operation on the first binarized image to form the second binarized image. The inversion operation, the opening operation, and the closing operation are all conventional image processing methods, which are also prior arts and will not be further elaborated here. It can be known here that after performing inversion, opening, and closing processing on the first binarized image, the working area in the formed second binarized image is not exactly the same as the non-working area in the first binarized image.

[0137] It should be noted that for the above steps M2 and M3, they can be executed successively in the order of steps M2 and M3, or they can be executed alternately. The change of the execution order does not affect the final output result.

[0138] For step M4, in the first preferred embodiment of the present invention, the first parameter is: the histogram peak maxH of the H-channel image; the third parameter is the ratio P1 of the number of edge pixel points corresponding to the non-working area within the rectangular contour in the edge image to the number of pixel points in the non-working area within the rectangular contour; the fourth parameter includes: the size DL of the rectangular contour, the number N3 of pixel points in the non-working area within the rectangular contour, and the ratio P2 of the number of pixel points in the working area within the rectangular contour to the number of pixel points in the non-working area; the size of the rectangular contour includes at least one of: the diagonal length XL of the rectangular contour, the length LL of the rectangular contour, and the width WL of the rectangular contour;

[0139] Correspondingly, configuring the preset parameter thresholds includes: a preset third peak H3, a preset fourth peak H4, a preset first ratio Q1, a preset first length value Le1, a preset third quantity M3, and a preset second ratio Q2;

[0140] and / or includes: a preset third peak H3, a preset fourth peak H4, a preset first ratio Q1, a preset first length value Le1, a preset fourth quantity M4, and a preset third ratio Q3;

[0141] and / or includes: a preset third peak H3, a preset fourth peak H4, a preset first ratio Q1, a preset first length value Le1, a preset fifth quantity M5, and a preset fourth ratio Q4;

[0142] and / or includes: a preset third peak H3, a preset fourth peak H4, a preset first ratio Q1, a preset first length value Le1, a preset sixth quantity M6, and a preset fifth ratio Q5;

[0143] and / or includes: a preset third peak H3, a preset fourth peak H4, a preset sixth ratio Q6, a preset first length value Le1, a preset sixth quantity M6, and a preset seventh ratio Q7;

[0144] Further, identifying the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter values includes:

[0145] c. If all of the following conditions are satisfied simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M3, P2 < Q2, then it is determined as a non-working area;

[0146] d. Or if all of the following conditions are satisfied simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M4, P2 < Q3, then it is determined as a non-working area;

[0147] e, or simultaneously satisfy: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M5, P2 < Q4, then it is determined as a non - working area;

[0148] f, or simultaneously satisfy: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M6, P2 < Q5, then it is determined as a non - working area;

[0149] g, or simultaneously satisfy: H3 < maxH < H4, P1 < Q6, DL > Le1, N3 > M6, P2 < Q7, then it is determined as a non - working area;

[0150] Among them, M3 < M4 < M5 < M6, Q2 < Q3 < Q4 < Q5 < Q7, Q1 > Q6.

[0151] Preferably, for the first preferred embodiment of implementing step M4, configure H3 = 1.56% * (L1 * W1), H4 = 2.08% * (L1 * W1), Q1 = 0.24, Le1 ∈ [50% * Lf1, 55% * Lf1], M3 = 18.2 * (L2 * W2), Q2 = 0.01;

[0152] and / or configure H3 = 1.56% * (L1 * W1), H4 = 2.08% * (L1 * W1), Q1 = 0.24, Le1 ∈ [50% * Lf1, 55% * Lf1], M4 = 26 * (L2 * W2), Q3 = 0.011;

[0153] and / or configure H3 = 1.56% * (L1 * W1), H4 = 2.08% * (L1 * W1), Q1 = 0.24, Le1 ∈ [50% * Lf1, 55% * Lf1], M5 = 31.3 * (L2 * W2), Q4 = 0.013;

[0154] and / or configure H3 = 1.56% * (L1 * W1), H4 = 2.08% * (L1 * W1), Q1 = 0.24, Le1 ∈ [50% * Lf1, 55% * Lf1], M6 = 36.5 * (L2 * W2), Q5 = 0.015;

[0155] and / or configure H3 = 1.56% * (L1 * W1), H4 = 2.08% * (L1 * W1), Q6 = 0.22, Le1 ∈ [50% * Lf1, 55% * Lf1], M6 = 36.5 * (L2 * W2), Q7 = 0.09;

[0156] Among them, L1 and W1 respectively represent the length and width of the first binary image, and Lf1 represents the size of the second binary image. Here, if the size DL of the rectangular contour is the diagonal length XL of the rectangular contour, then Lf1 represents the diagonal length of the second binary image; if the size DL of the rectangular contour is the length LL of the rectangular contour, then Lf1 represents the length of the second binary image; if the size DL of the rectangular contour is the width WL of the rectangular contour, then Lf1 represents the width of the second binary image; L2 and W2 respectively represent the length and width of the second binary image.

[0157] Preferably, the value configuration of Le1 is 52.5 * Lf1.

[0158] In the first preferred embodiment of the implementation step M4, the size DL of the rectangular contour and the number N3 of non-working area pixel points within the rectangular contour can both reflect the size of the area enclosed by the rectangular contour.

[0159] In addition, in this example, through the comparison of multiple groups of parameters of rules c to f, it can be known that: when the number N3 of non-working area pixel points within the rectangular contour is larger, it indicates that the non-working area range is larger, so the requirement for reducing the ratio P2 of the number of working area pixel points to the number of non-working area pixel points within the rectangular contour is reduced.

[0160] Through the comparison of multiple groups of parameters of rules f and g, it can be known that: when the ratio P1 of the number of edge pixel points corresponding to the non-working area within the rectangular contour in the edge image to the number of pixel points in the non-working area within the rectangular contour is smaller, it indicates that the roughness is lower and the image is smoother. At this time, the condition for the size of the working area within the rectangular contour is appropriately relaxed, that is, when the value of the ratio P2 of the number of working area pixel points to the number of non-working area pixel points within the rectangular contour is larger, the robot can still be in the non-lawn area. Because in this implementation, when the roughness of the working area is higher than that of the non-working area, the roughness is relatively low and the probability of the non-working area is relatively large.

[0161] For step M4, in the second preferred embodiment of the present invention, the first parameter is: the histogram peak maxH of the H-channel image; the third parameter is the ratio P1 of the number of edge pixel points corresponding to the non-working area within the rectangular contour in the edge image to the number of pixel points in the non-working area within the rectangular contour; the fourth parameters include: the size DL of the rectangular contour, the number N3 of non-working area pixel points within the rectangular contour, the ratio P2 of the number of working area pixel points to the number of non-working area pixel points within the rectangular contour, and the ratio P3 of the number of non-working area pixel points corresponding to the working area in the first binary image within the rectangular contour to the number of pixel points in the non-working area within the rectangular contour; the size of the rectangular contour includes at least one of the diagonal length XL of the rectangular contour, the length LL of the rectangular contour, and the width WL of the rectangular contour.

[0162] Correspondingly, configuring the preset parameter thresholds includes: preset fourth peak value H4, preset fifth peak value H5, preset eighth ratio Q8, preset first length value Le1, preset seventh quantity M7, preset ninth ratio Q9, and preset tenth ratio Q10;

[0163] Further, identifying the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter, and the preset parameter values includes:

[0164] h, if all of the following are satisfied simultaneously: H4 < maxH < H5, P1 < Q8, DL > Le1, N3 > M7, P2 < Q9, P3 < Q10, then it is determined as a non-working area.

[0165] Preferably, for the second preferred implementation manner of implementing step M4, configure H4 = 2.08% * (L1 * W1), H5 = 2.86% * (L1 * W1), Q8 = 0.27, Le1 ∈ [50% * Lf1, 55% * Lf1], M7 = 17.2 * (L2 * W2), Q9 = 0.01, Q10 = 0.27; where L1 and W1 respectively represent the length and width of the first binary image, Lf1 represents the size of the second binary image. Here, if the rectangle contour size DL is the diagonal length XL of the rectangle contour, then Lf1 represents the diagonal length of the second binary image; if the rectangle contour size DL is the length LL of the rectangle contour, then Lf1 represents the length of the second binary image; if the rectangle contour size DL is the width WL of the rectangle contour, then Lf1 represents the width of the second binary image; L2 and W2 respectively represent the length and width of the second binary image.

[0166] Preferably, the value of Le1 is configured as 52.5 * Lf1.

[0167] For step M4, in the third preferred implementation manner of the present invention, the first parameter is: the histogram peak value maxH of the H-channel image; the third parameter is the ratio P1 of the number of edge pixel points corresponding to the non-working area within the rectangle contour in the edge image to the number of pixel points in the non-working area within the rectangle contour; the fourth parameter includes: the rectangle contour size DL, and the ratio P2 of the number of pixel points in the working area to the number of pixel points in the non-working area within the rectangle contour; the ratio P3 of the number of pixel points corresponding to the non-working area within the rectangle contour in the working area of the first binary image to the number of pixel points in the non-working area within the rectangle contour; the rectangle contour size includes at least one of the diagonal length XL of the rectangle contour, the length LL of the rectangle contour, and the width WL of the rectangle contour;

[0168] The preset parameter thresholds include: preset fifth peak value H5, preset eleventh ratio Q11, preset first length value Le1, preset twelfth ratio Q12, preset thirteenth ratio Q13;

[0169] and / or preset fifth peak value H5, preset fourteenth ratio Q14, preset first length value Le1, preset fifteenth ratio Q15, preset thirteenth ratio Q13;

[0170] and / or preset fifth peak value H5, preset sixteenth ratio Q16, preset first length value Le1, preset seventeenth ratio Q17, preset thirteenth ratio Q13.

[0171] Further, identifying the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter value includes:

[0172] i. If all of the following conditions are satisfied simultaneously: maxH≥H5, P1<Q11, DL>Le1, P2<Q12, P3<Q13, then it is determined as a non - working area;

[0173] j. Or if all of the following conditions are satisfied simultaneously: maxH≥H5, P1<Q14, DL>Le1, P2<Q15, P3≥Q13, then it is determined as a non - working area;

[0174] k. If all of the following conditions are satisfied simultaneously: maxH≥H5, P1<Q16, DL>Le1, P2<Q17, P3≥Q13, then it is determined as a non - working area;

[0175] Among them, Q11>Q14>Q16, Q12<Q15<Q17.

[0176] Preferably, configure H5 = 2.86%*(L1*W1), Q11 = 0.24, Le1∈[50%*Lf1, 55%*Lf1], Q12 = 0.11, Q13 = 0.4;

[0177] and / or configure H5 = 2.86%*(L1*W1), Q14 = 0.195, Le1∈[50%*Lf1, 55%*Lf1], Q15 = 0.2, Q13 = 0.4;

[0178] and / or configure H5 = 2.86%*(L1*W1), Q16 = 0.16, Le1∈[50%*Lf1, 55%*Lf1], Q17 = 0.23, Q13 = 0.4;

[0179] Among them, L1 and W1 respectively represent the length and width of the first binary image, and Lf1 represents the size of the second binary image. Here, if the size DL of the rectangular contour is the diagonal length XL of the rectangular contour, then Lf1 represents the diagonal length of the second binary image; if the size DL of the rectangular contour is the length LL of the rectangular contour, then Lf1 represents the length of the second binary image; if the size DL of the rectangular contour is the width WL of the rectangular contour, then Lf1 represents the width of the second binary image; L2 and W2 respectively represent the length and width of the second binary image.

[0180] Preferably, the value of Le1 is configured as 52.5 * Lf1.

[0181] In addition, in this example, through the comparison of multiple groups of parameters from rules h to k, it can be known that: the larger the value of the ratio P3 of the number of non - working area pixel points within the rectangular contour corresponding to the working area on the first binary image to the number of pixel points in the non - working area within the rectangular contour, and the higher the histogram peak maxH of the H - channel image, it indicates that the working area contained in the image is larger. Correspondingly, the value of the ratio P2 of the number of working area pixel points to the number of non - working area pixel points within the rectangular contour is also larger. At this time, to determine whether the robot is in a non - lawn area, it can be judged by the size of the ratio P1 of the number of edge pixel points corresponding to the non - working area within the rectangular contour in the edge image to the number of pixel points in the non - working area within the rectangular contour. The smaller the value of P1, the greater the probability that the robot is in a non - working area.

[0182] Preferably, for the method of identifying the working area based on the image in the above first and second embodiments, if it is confirmed that the robot is currently in a non - working area, then the method further includes: obtaining the rectangular contour coordinate parameter value Y based on the second binary image; when Y is greater than the preset coordinate value, driving the robot to execute the obstacle avoidance logic; where the value range of the preset coordinate value is [58% * (L2 * W2), 66% * (L2 * W2)], and L2 and W2 respectively represent the length and width of the second binary image.

[0183] Preferably, the value of Y is 62.5 * (L2 * W2).

[0184] Here, the coordinate parameter value Y can be the y - axis coordinate of the lower - right corner of the rectangular contour, or the central position coordinate of the rectangular contour, or the average value of the pixel coordinate Y values of the non - lawn area in the contour, etc., which will not be elaborated further here.

[0185] It should be noted that in the method for identifying the working area based on image recognition in the first embodiment, before obtaining the rectangular contour coordinate parameter value Y, the following steps need to be performed first: Invert the first binary image to form a second binary image; the second binary image includes a working area and a non-working area with different pixel values; based on the second binary image, obtain the smallest rectangular contour enclosing the non-working area of the second binary image; preferably, the second binary image is formed by sequentially performing inversion, opening, and closing operations on the first binary image. For the method for identifying the working area based on image recognition in the second embodiment, it has been converted into a second binary image and the smallest rectangular contour enclosing the non-working area of the second binary image has been obtained in the foregoing steps; therefore, it can directly obtain the rectangular contour coordinate parameter value Y on the basis of the second binary image, and no further elaboration will be made here.

[0186] In one embodiment of the present invention, a robot is further provided, including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the method for identifying the working area based on image recognition in any of the foregoing embodiments are implemented.

[0187] Combined with Figure 4 As shown, a system for identifying the working area based on image recognition is provided, and the system includes: an acquisition module 100, a conversion module 200, and an analysis module 300.

[0188] For the system for identifying the working area based on image recognition in the first embodiment of the present invention, the acquisition module 100 is used to acquire an original image; the conversion module 200 is used to separate the H-channel image from the original image; perform binary processing on the original image to form a first binary image; the first binary image includes a working area and a non-working area with different pixel values; the analysis module 300 is used to count the histogram of the H-channel image and obtain a first parameter representing the peak value of the histogram; based on the first binary image, count a second parameter representing the size of the non-working area in the first binary image; identify the working area according to the magnitude relationship between the first parameter, the second parameter, and a preset parameter threshold.

[0189] Furthermore, the acquisition module 100 in the system for identifying the working area based on image recognition in the first embodiment is used to implement step S1; the conversion module 200 is used to implement step S2; the analysis module 300 is used to implement steps S3 and S4, and perform calculations for driving the robot to execute an obstacle avoidance logic; those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described system can refer to the corresponding process in the foregoing method embodiment, and will not be elaborated here.

[0190] For the system for identifying a working area based on image recognition provided in the second embodiment of the present invention, an acquisition module 100 is configured to acquire an original image; a conversion module 200 separates a V-channel image from the original image, performs edge extraction on the V-channel image to form an edge image, and separates an H-channel image from the HSV image, and statistically analyzes a histogram of the H-channel image; performs binarization processing on the original image to form a first binarized image, and inverts the first binarized image to form a second binarized image; both the first binarized image and the second binarized image include a working area and a non-working area with different pixel values; based on the second binarized image, acquires a minimum rectangular contour that encloses the non-working area of the second binarized image; an analysis module 300 is configured to acquire a first parameter representing the peak value of the histogram of the H-channel image based on the H-channel image; statistically analyzes a third parameter representing the roughness of the edge image within the range of the non-working area enclosed by the rectangular contour corresponding to the edge image; acquires a fourth parameter representing the attributes of the rectangular contour based on the second binarized image; and identifies the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and a preset parameter value.

[0191] Further, the acquisition module 100 in the system for identifying a working area based on image recognition in the second embodiment is configured to implement step M1; the conversion module 200 is configured to implement step M2; the analysis module 300 is configured to implement steps M3 and M4, and perform calculations for driving a robot to execute an obstacle avoidance logic; those skilled in the art can clearly understand that for the sake of convenience and brevity of description, the specific working process of the above-described system can refer to the corresponding process in the foregoing method embodiment, and will not be elaborated herein.

[0192] In summary, the method, system and robot for identifying a working area based on image recognition of the present invention can accurately identify the working area through image recognition, save costs, and improve the performance of the robot.

[0193] In several embodiments provided in the present application, it should be understood that the disclosed modules, systems and methods can all be implemented in other ways. The system embodiments described above are only illustrative, and the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0194] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they may be located in one place, or may be distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0195] In addition, in each embodiment of the present application, each functional module can be integrated into one parsing module, or each module can exist physically alone, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware, or in the form of a combination of hardware and software functional modules.

[0196] 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 foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. A method for identifying a working area based on image recognition, characterized in that, The method includes: Obtain the original image; Separate the H-channel image from the original image; Perform binarization processing on the original image to form a first binarized image; the first binarized image includes a working area and a non-working area with different pixel values; Statistically analyze the histogram of the H-channel image and obtain a first parameter representing the peak value of the histogram; Based on the first binarized image, statistically analyze a second parameter representing the size of the non-working area in the first binarized image; Identify the working area according to the magnitude relationship between the first parameter, the second parameter and a preset parameter threshold; The first parameter is: the peak value maxH of the histogram of the H-channel image; The second parameter is: the number of pixel points N1 in the non-working area of the first binarized image; The preset parameter threshold includes: a preset first value M1 and a preset first peak value H1, and / or includes a preset second value M2 and a preset second peak value H2; Identifying the working area according to the magnitude relationship between the first parameter, the second parameter and the preset parameter threshold includes: If both maxH > H1 and N1 > M1 are satisfied, it is determined as a non-working area; Or if both maxH ≤ H2 and N1 > M2 are satisfied, it is determined as a non-working area; Wherein, H1 = H2 and M1 < M2.

2. The method for identifying a working area based on image recognition according to claim 1, characterized in that, The method further includes: Invert the first binarized image to form a second binarized image; the second binarized image includes a working area and a non-working area with different pixel values; Based on the second binarized image, obtain the smallest rectangular contour enclosing the non-working area in the second binarized image; Based on the second binarized image, obtain the rectangular contour coordinate parameter value Y; If it is confirmed that the robot is currently in a non-working area and Y is greater than the preset coordinate value, then drive the robot to execute the obstacle avoidance logic; Wherein, the value range of the preset coordinate value is [58%*(L2*W2), 66%*(L2*W2)], and L2 and W2 respectively represent the length and width of the second binarized image.

3. A method for identifying a working area based on image recognition, characterized in that, The method includes: Obtain the original image; Separate the V-channel image from the original image, perform edge extraction on the V-channel image to form an edge image, and separate the H-channel image from the HSV image, and statistically analyze the histogram of the H-channel image; Perform binarization processing on the original image to form a first binarized image, and invert the first binarized image to form a second binarized image; both the first binarized image and the second binarized image include a working area and a non-working area with different pixel values; Based on the second binarized image, obtain the smallest rectangular contour enclosing the non-working area in the second binarized image; Based on the H-channel image, obtain a first parameter representing the peak value of the histogram of the H-channel image; Based on the edge image, statistically analyze a third parameter representing the roughness of the edge image within the range of the non-working area enclosed by the rectangular contour; Obtain a fourth parameter characterizing the properties of the rectangular contour based on the second binary image; Identify the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter threshold; The first parameter is: the histogram peak maxH of the H-channel image; the third parameter is the ratio P1 of the number of edge pixel points in the edge image corresponding to the non-working area within the rectangular contour to the number of pixel points in the non-working area within the rectangular contour; the fourth parameter includes: the size DL of the rectangular contour, the number N3 of pixel points in the non-working area within the rectangular contour, and the ratio P2 of the number of pixel points in the working area within the rectangular contour to the number of pixel points in the non-working area; the size of the rectangular contour includes at least one of: the diagonal length XL of the rectangular contour, the length LL of the rectangular contour, and the width WL of the rectangular contour; Configuring the preset parameter threshold includes: a preset third peak H3, a preset fourth peak H4, a preset first ratio Q1, a preset first length value Le1, a preset third quantity M3, and a preset second ratio Q2; and / or includes; a preset third peak H3, a preset fourth peak H4, a preset first ratio Q1, a preset first length value Le1, a preset fourth quantity M4, and a preset third ratio Q3; and / or includes; a preset third peak H3, a preset fourth peak H4, a preset first ratio Q1, a preset first length value Le1, a preset fifth quantity M5, and a preset fourth ratio Q4; and / or includes; a preset third peak H3, a preset fourth peak H4, a preset first ratio Q1, a preset first length value Le1, a preset sixth quantity M6, and a preset fifth ratio Q5; and / or includes; a preset third peak H3, a preset fourth peak H4, a preset sixth ratio Q6, a preset first length value Le1, a preset sixth quantity M6, and a preset seventh ratio Q7; and / or includes: a preset fourth peak H4, a preset fifth peak H5, a preset eighth ratio Q8, a preset first length value Le1, a preset seventh quantity M7, a preset ninth ratio Q9, and a preset tenth ratio Q10; and / or includes: a preset fifth peak H5, a preset eleventh ratio Q11, a preset first length value Le1, a preset twelfth ratio Q12, and a preset thirteenth ratio Q13; and / or includes: a preset fifth peak H5, a preset fourteenth ratio Q14, a preset first length value Le1, a preset fifteenth ratio Q15, and a preset thirteenth ratio Q13; and / or includes: a preset fifth peak H5, a preset sixteenth ratio Q16, a preset first length value Le1, a preset seventeenth ratio Q17, and a preset thirteenth ratio Q13; Identifying the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter threshold includes any of the following methods: Method 1: If all of the following are satisfied simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M3, P2 < Q2, then it is determined as a non-working area; Or if all of the following are satisfied simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M4, P2 < Q3, then it is determined as a non-working area; Or simultaneously satisfy: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M5, P2 < Q4, then it is determined as a non - working area; Or simultaneously satisfy: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M6, P2 < Q5, then it is determined as a non - working area; Or simultaneously satisfy: H3 < maxH < H4, P1 < Q6, DL > Le1, N3 > M6, P2 < Q7, then it is determined as a non - working area; Among them, M3 < M4 < M5 < M6, Q2 < Q3 < Q4 < Q5 < Q7, Q1 > Q6; Method 2: If simultaneously satisfy: H4 < maxH < H5, P1 < Q8, DL > Le1, N3 > M7, P2 < Q9, P3 < Q10, then it is determined as a non - working area; Method 3: If simultaneously satisfy: maxH ≥ H5, P1 < Q11, DL > Le1, P2 < Q12, P3 < Q13, then it is determined as a non - working area; Or simultaneously satisfy: maxH ≥ H5, P1 < Q14, DL > Le1, P2 < Q15, P3 ≥ Q13, then it is determined as a non - working area; Or simultaneously satisfy: maxH ≥ H5, P1 < Q16, DL > Le1, P2 < Q17, P3 ≥ Q13, then it is determined as a non - working area; Among them, Q11 > Q14 > Q16, Q12 < Q15 < Q17.

4. A system for identifying a working area based on image recognition, characterized in that the system includes: an acquisition module, configured to acquire an original image; a conversion module, configured to separate an H - channel image from the original image; perform binarization processing on the original image to form a first binarized image; the first binarized image includes a working area and a non - working area with different pixel values; an analysis module, configured to count the histogram of the H - channel image and obtain a first parameter representing the peak value of the histogram; statistically obtain a second parameter representing the size of the non - working area in the first binarized image based on the first binarized image; identify the working area according to the magnitude relationship between the first parameter, the second parameter and a preset parameter threshold; wherein, the first parameter is: the peak value maxH of the histogram of the H - channel image; the second parameter is: the number of non - working area pixel points N1 in the first binarized image; the preset parameter threshold includes: a preset first value M1 and a preset first peak value H1, and / or includes a preset second value M2 and a preset second peak value H2; Identifying the working area according to the magnitude relationship between the first parameter, the second parameter and the preset parameter threshold includes: If simultaneously satisfy: maxH > H1, N1 > M1, then it is determined as a non - working area; Or simultaneously satisfy: maxH ≤ H2, N1 > M2, then it is determined as a non - working area.

5. A system for identifying a working area based on image recognition, characterized in that the system includes: an acquisition module, configured to acquire an original image; a conversion module, configured to separate a V - channel image from the original image, perform edge extraction on the V - channel image to form an edge image, and separate an H - channel image from the HSV image and count the histogram of the H - channel image; Perform binarization processing on the original image to form a first binarized image, and invert the first binarized image to form a second binarized image; both the first binarized image and the second binarized image include a working area and a non-working area with different pixel values; Based on the second binarized image, obtain the smallest rectangular contour that encloses the non-working area of the second binarized image; An analysis module for obtaining a first parameter representing the peak value of the histogram of the H-channel image based on the H-channel image; Statistically obtain a third parameter representing the roughness of the edge image within the range of the non-working area enclosed by the rectangular contour corresponding to the edge image; Obtain a fourth parameter representing the attributes of the rectangular contour based on the second binarized image; Identify the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter threshold; The first parameter is: the peak value maxH of the histogram of the H-channel image; the third parameter is the ratio P1 of the number of edge pixel points in the edge image corresponding to the non-working area within the rectangular contour to the number of pixel points in the non-working area within the rectangular contour; the fourth parameter includes: the size DL of the rectangular contour, the number N3 of pixel points in the non-working area within the rectangular contour, and the ratio P2 of the number of pixel points in the working area within the rectangular contour to the number of pixel points in the non-working area; the size of the rectangular contour includes at least one of: the diagonal length XL of the rectangular contour, the length LL of the rectangular contour, and the width WL of the rectangular contour; Configuring the preset parameter threshold includes: a preset third peak value H3, a preset fourth peak value H4, a preset first ratio Q1, a preset first length value Le1, a preset third quantity M3, and a preset second ratio Q2; And / or include; a preset third peak value H3, a preset fourth peak value H4, a preset first ratio Q1, a preset first length value Le1, a preset fourth quantity M4, and a preset third ratio Q3; And / or include; a preset third peak value H3, a preset fourth peak value H4, a preset first ratio Q1, a preset first length value Le1, a preset fifth quantity M5, and a preset fourth ratio Q4; And / or include; a preset third peak value H3, a preset fourth peak value H4, a preset first ratio Q1, a preset first length value Le1, a preset sixth quantity M6, and a preset fifth ratio Q5; And / or include; a preset third peak value H3, a preset fourth peak value H4, a preset sixth ratio Q6, a preset first length value Le1, a preset sixth quantity M6, and a preset seventh ratio Q7; And / or include: a preset fourth peak value H4, a preset fifth peak value H5, a preset eighth ratio Q8, a preset first length value Le1, a preset seventh quantity M7, a preset ninth ratio Q9, and a preset tenth ratio Q10; And / or include: a preset fifth peak value H5, a preset eleventh ratio Q11, a preset first length value Le1, a preset twelfth ratio Q12, and a preset thirteenth ratio Q13; And / or include: a preset fifth peak value H5, a preset fourteenth ratio Q14, a preset first length value Le1, a preset fifteenth ratio Q15, and a preset thirteenth ratio Q13; and / or including: a preset fifth peak value H5, a preset sixteenth proportion Q16, a preset first length value Le1, a preset seventeenth proportion Q17, a preset thirteenth proportion Q13; Identifying the working area according to the magnitude relationship between the first parameter, the third parameter, the fourth parameter and the preset parameter threshold, including any one of the following methods: Method 1: If all of the following conditions are satisfied simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M3, P2 < Q2, then it is determined as a non-working area; Or if all of the following conditions are satisfied simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M4, P2 < Q3, then it is determined as a non-working area; Or if all of the following conditions are satisfied simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M5, P2 < Q4, then it is determined as a non-working area; Or if all of the following conditions are satisfied simultaneously: H3 < maxH < H4, P1 < Q1, DL > Le1, N3 > M6, P2 < Q5, then it is determined as a non-working area; Or if all of the following conditions are satisfied simultaneously: H3 < maxH < H4, P1 < Q6, DL > Le1, N3 > M6, P2 < Q7, then it is determined as a non-working area; wherein, M3 < M4 < M5 < M6, Q2 < Q3 < Q4 < Q5 < Q7, Q1 > Q6; Method 2: If all of the following conditions are satisfied simultaneously: H4 < maxH < H5, P1 < Q8, DL > Le1, N3 > M7, P2 < Q9, P3 < Q10, then it is determined as a non-working area; Method 3: If all of the following conditions are satisfied simultaneously: maxH ≥ H5, P1 < Q11, DL > Le1, P2 < Q12, P3 < Q13, then it is determined as a non-working area; Or if all of the following conditions are satisfied simultaneously: maxH ≥ H5, P1 < Q14, DL > Le1, P2 < Q15, P3 ≥ Q13, then it is determined as a non-working area; Or if all of the following conditions are satisfied simultaneously: maxH ≥ H5, P1 < Q16, DL > Le1, P2 < Q17, P3 ≥ Q13, then it is determined as a non-working area; wherein, Q11 > Q14 > Q16, Q12 < Q15 < Q17.

6. A robot, including a memory and a processor, the memory stores a computer program, and is characterized in that when the processor executes the computer program, it implements the steps of the method for identifying the working area based on image recognition according to any one of claims 1 - 3.

Images