CN107463167A - Automatic walking equipment and target area recognition method - Google Patents

Abstract

The invention relates to a method for identifying a target area of walking of automatic walking equipment, which is characterized by comprising the following steps of: s10, acquiring an image of the walking target area of the automatic walking equipment; s20, dividing the image into a plurality of cells, wherein each cell has at least one adjacent cell; s30, identifying whether a target area corresponding to the cell is a working area or not according to the color information of the designated pixel in the cell and the texture characteristic value of the cell, and obtaining an identification result; s60, for each cell, the recognition result obtained by the step S30 is changed or maintained according to the recognition result of the adjacent cell.

Description

Automatic walking equipment and target area recognition method

technical field

The invention relates to an automatic running device and a method for identifying a target area by the automatic running device.

Background technique

With the continuous advancement of computer technology and artificial intelligence technology, automatic walking equipment similar to intelligent robots has slowly entered people's lives. Companies such as Samsung and Electrolux have developed fully automatic vacuum cleaners and have put them into the market. This kind of fully automatic vacuum cleaner is usually small in size and integrates environmental sensors, self-driving system, vacuum system, battery and charging system. It can cruise indoors by itself without manual control, and automatically return to the docking station when the energy is low, docking and charging. Then go on cruise vacuuming. At the same time, companies such as Haskovana have developed similar smart lawn mowers, which can automatically mow and recharge the user's lawn without user intervention. Since this automatic mowing system does not need to invest in energy management after one setting, it liberates users from boring and time-consuming housework such as cleaning and lawn maintenance, so it is very popular.

The walking area of the existing automatic lawn mower is generally set by a physical boundary line, such as a wire or a fence, and the automatic lawn mower detects the physical boundary line to determine the walking area. The process of boundary wiring is cumbersome, time-consuming and laborious, and there may be non-grass areas within the boundary line, or areas that need to be mowed outside the boundary line. The method of using a physical boundary line is inflexible and inconvenient.

However, the use of electronic means to identify and determine the solution to the walking area, due to the diversity of the walking area, using existing methods often produces noise, the accuracy of identifying the walking area is very low, which affects the judgment of the automatic walking equipment, and it is easy to leave The walking area will have a certain impact on the normal work of the automatic walking equipment.

Therefore, it is necessary to improve the existing technical means, so that the walking area can be identified more accurately, so as to facilitate the automatic walking equipment to work.

Contents of the invention

In view of this, one of the objectives of the present invention is to provide a method for accurately identifying a target area and controlling an automatic walking device to walk correspondingly according to the recognition result, and an automatic running device using the method.

In order to achieve the above object, the technical solution adopted in the present invention is: a method for controlling the walking of automatic walking equipment, which is characterized in that it includes the following steps: S10, acquiring the image of the walking target area of the automatic walking equipment; S20, dividing the image into into several cells, each cell having at least one adjacent cell; S30, identifying the target area corresponding to the cell according to the color information of the specified pixel in the cell and the texture feature value of the cell Whether it is a working area, and obtain the recognition result; S40, divide the image into several sub-image blocks, each sub-image block includes several adjacent cells, and judge the Whether the target area corresponding to the sub-image block is a walkable area, and obtain a judgment result; S50, according to the judgment result, control the walking direction of the automatic walking device.

Preferably, the method further includes, after obtaining the recognition result, for each cell, adjusting the recognition result of the cell according to the recognition results of the cell and its adjacent cells.

Preferably, the method further includes the following steps: S61, selecting a cell, and obtaining its recognition result; S62, counting the number of adjacent cells with the same recognition result as in step S61; S63, calculating the number of cells in step S62 The ratio of the number of cells in the total number of adjacent cells; S64, if the ratio exceeds or reaches the preset value, then maintain the recognition result of the cell specified in step S61 unchanged; if the ratio is less than the preset value, then change The recognition result of the cell specified in step S61, wherein the fourth preset value is greater than or equal to 50%.

Preferably, the adjacent cells include cells adjacent to the selected cell horizontally and vertically.

Preferably, the adjacent cells further include cells adjacent to the selected cell in a direction forming an angle of 45 degrees with the horizontal direction and the vertical direction.

Preferably, the method further includes the following steps: S66. Select a cell, and obtain the reliability Y1 of the cell for its recognition result, where the reliability Y1 is a value between 0% and 100%; S67. Calculate 1-Y1, and mark the result as N1; S68, obtain the reliability Ya, Yb... of the recognition results of all adjacent cells of the selected cell, and the reliability Ya, Yb... is between 0 and 100% A numerical value; S69, calculate 1-Ya, 1-Yb..., and mark the result as Na, Nb...; S70, weight and sum Ya, Yb... to obtain the weighted sum Y2, and sum the weighted sum of Nb, Nc... Obtaining the weighted sum is N2, wherein the weighting coefficients are all the same; S71, respectively calculating the results of Y1+αN1 and Y2+αN2 and comparing their sizes, wherein α is a coefficient; S72, if the result of Y1+αN1 is greater than or equal to Y2+αN2 If the result of Y1+αN1 is smaller than the result of Y2+αN2, then the recognition result of the designated cell is changed.

Preferably, the step S40 further includes the following steps: S41. Divide the image into several sub-image blocks, obtain the number of cells contained in each sub-image block, and mark it as B; collect the number of cells in the sub-image blocks The recognition result of the cell, count the number of cells whose recognition result is the working area, and mark it as A; if A:B is less than the third preset value, then it is judged that the target area corresponding to the sub-image block is not available a walking area; otherwise, it is determined that the target area corresponding to the sub-image block is a walking area.

Preferably, the method further includes continuously shooting the same target area to form multiple frames of images, judging whether the target area corresponding to the sub-image block is a walkable area according to the judgment result of the same sub-image block in each frame of image, and obtaining the judgment result.

Preferably, the method further includes the following steps: S81, taking continuous shots of the same target area to form multiple frames of images; S82, selecting a sub-image block in one of the frames of images, and obtaining the judgment result through step S40; S83, setting the initial , and calculate the parameter value according to the judgment result obtained in step S82, if the judgment result is a walkable area, then add the first parameter associated with the judgment result to the initial parameter value to become the current parameter value; if it is judged The result is not a walkable area, then keep the parameter value unchanged; S84, select the next frame image, and calculate the current parameter value according to the judgment result obtained in step S82, if the judgment result is a walkable area, then in Adding the first parameter associated with the judgment result to the current parameter value becomes a new current parameter value; if the judgment result is not a walkable area, then keep the current parameter value unchanged; S85, compare the current parameter value with The size of the threshold, if the current parameter value is greater than or equal to the threshold, it is determined that the target area corresponding to the sub-image block is a walkable area.

Preferably, the step S84 further includes: after the next frame of image is selected and before the operation is performed on the current parameter value, the current parameter value minus a preset second parameter, and the second parameter is less than the first parameter.

Preferably, the sub-image blocks include three sub-image blocks of a middle part, a left part and a right part, respectively corresponding to a middle area, a left area and a right area of the target area.

In order to achieve the above object, another technical solution adopted by the present invention is: a kind of self-propelled equipment, characterized in that: comprising a housing, an image acquisition device located on the housing, the image acquisition device is used to photograph the target area and Generate images, drive the walking module of the automatic walking equipment to walk, connect the image acquisition device and the walking module to control the main control module of the automatic walking equipment, wherein the main control module includes a division unit, an identification unit, and a judgment unit and a control unit, the dividing unit divides the image into several cells, and transmits the division result to the recognition unit, and the recognition unit recognizes whether the target area corresponding to the cell is a working area, and sends The recognition result is passed to the judging unit, which judges whether the corresponding area of the sub-image block containing several cells is a walkable area, and passes the judging result to the control unit, and the control unit controls the walking direction of the walking module according to the judging result.

Preferably, the main control module further includes a correction unit, for each cell, the correction unit adjusts the recognition result of the cell according to the recognition results of the cell and its adjacent cells.

Preferably, the adjacent cells include cells adjacent to the selected cell horizontally and vertically.

Preferably, the adjacent cells further include cells adjacent to the selected cell in a direction forming an angle of 45 degrees with the horizontal direction and the vertical direction.

Preferably, the judging unit further includes a sub-image block dividing unit, the sub-image block dividing unit divides the image into several sub-image blocks, and the judging unit judges the corresponding sub-image according to the recognition results of the cells contained in the sub-image blocks Whether the block is a walkable area.

Preferably, the sub-image blocks include three sub-image blocks in the middle, left and right.

Preferably, the main control module further includes a recording unit that records initial parameter values, the image acquisition device continuously captures the same target area to form multiple frames of images, and the judgment unit performs a check on the same sub-image block in each frame of image Making a judgment and obtaining a judgment result, the recording unit calculates the parameter value according to the judgment result, and when the parameter value is greater than or equal to a threshold, it is determined that the target area corresponding to the sub-image block is a walkable area.

Compared with the prior art, the beneficial effects of the present invention are as follows: the image of the target area is divided into cells, each cell is recognized microscopically, and the recognition results of multiple cells are comprehensively identified macroscopically, so that Improve the accuracy of identifying the target area, which will help the automatic walking equipment to walk more accurately in the target area.

Another object of the present invention is to provide a method for accurately identifying a target area and an automatic walking device using the method.

In order to achieve the above object, a kind of technical solution adopted by the present invention is: a kind of recognition method about the target area that automatic walking equipment walks, it is characterized in that, described recognition method comprises the following steps: S10, obtain the information about the automatic walking The image of the device walking target area; S20, divide the image into several cells, each cell has at least one adjacent cell; S30, according to the color information of the specified pixel in the cell and the color information of the cell The texture feature value identifies whether the target area corresponding to the cell is a working area, and obtains an identification result; S60, for each cell, change or maintain the identification result obtained in step S30 according to the identification results of adjacent cells.

Preferably, the step S60 further includes the following steps: S61, designate a cell, and obtain its recognition result; S62, count the number of adjacent cells with the same recognition result as in step S61; S63, calculate the number of cells in step S62 The ratio of the quantity to the total number of adjacent cells; S64, if the ratio exceeds or reaches the fourth preset value, then maintain the recognition result of the cell specified in step S61; if the ratio is less than the fourth preset value , then change the recognition result of the cell specified in step S61, wherein the fourth preset value is greater than or equal to 50%; S65, execute the above steps S61-S64 for all cells.

Preferably, the adjacent cells include cells adjacent to the cell horizontally and vertically.

Preferably, the adjacent cells further include cells adjacent to the cells in a direction forming an angle of 45 degrees with the horizontal direction and the vertical direction.

Preferably, the step S60 further includes the following steps: S66, designate a cell, and obtain the reliability Y1 of the cell for its recognition result, the reliability Y1 is a value between 0% and 100%; S67, calculate 1-Y1, and mark the result as N1; S68, obtain the reliability Ya, Yb... of the recognition results of all adjacent cells of the specified cell, and the reliability Ya, Yb... is between 0 and 100% A numerical value; S69, calculate 1-Ya, 1-Yb..., and mark the result as Na, Nb...; S70, weight and sum Ya, Yb... to obtain a weighted sum Y2, and obtain a weighted sum of Nb, Nc... Weighted sum N2, wherein, the weighted coefficients here are all the same; S71, respectively calculate the results of Y1+αN1 and Y2+αN2 and compare their sizes, where α is a coefficient, if Y1+αN1 is greater than or equal to Y2+αN2, then maintain all The recognition result of the specified cell remains unchanged, if the result of Y1+αN1 is less than the result of Y2+αN2, then change the recognition result of the specified cell; S72, perform the above steps S66 to S71 for all cells, until The recognition results of all cells will not change.

In order to achieve the above object, another technical solution adopted by the present invention is: a kind of automatic walking equipment, characterized in that it includes a housing and an image acquisition device located on the housing, and the image acquisition device is used to photograph the target area And generate an image, drive the walking module of the automatic walking equipment to walk, connect the image acquisition device and the walking module to control the main control module of the automatic walking equipment, wherein the main control module divides the unit, the identification unit and the correction unit , the division unit divides the image into several cells, the identification unit identifies whether the target area corresponding to the cell is a working area, and transmits the identification result to the correction unit, and the correction unit for each The cell changes or maintains the recognition result obtained by the recognition unit according to the recognition result of the adjacent cells.

Preferably, the cells adjacent to the cell include cells adjacent to the cell horizontally and vertically.

Preferably, the cells adjacent to the cell also include cells adjacent to the cell in a direction forming an angle of 45 degrees with the horizontal direction and the vertical direction.

Compared with the prior art, the beneficial effects of the present invention are as follows: the image of the target area is divided into cells, each cell is recognized microscopically, and the recognition results of multiple cells are macroscopically corrected, thereby improving Accuracy in identifying target areas.

Description of drawings

The purpose, technical solutions and beneficial effects of the present invention described above can be clearly obtained through the following detailed description of the specific embodiments capable of realizing the present invention, combined with the description of the accompanying drawings.

The same reference numerals and symbols are used in the drawings and the specification to denote the same or equivalent elements.

Fig. 1 is a schematic diagram of an automatic walking device walking in a target area according to an embodiment of the present invention.

Fig. 2 is a schematic diagram of a target area photographed by the automatic walking device in Fig. 1 .

Fig. 3 is a schematic diagram of dividing target areas by the automatic walking device in Fig. 1 .

Fig. 4 is a schematic diagram of various modules of the self-propelled device in Fig. 1 .

Fig. 5 is a schematic flowchart of a method for controlling the walking of an automatic walking device according to an embodiment of the present invention.

FIG. 6 is a schematic flowchart of step S60 between step S30 and step S40 in an embodiment of the present invention.

FIG. 7 is a schematic flowchart of step S60 between step S30 and step S40 in another embodiment of the present invention.

FIG. 8 is a detailed flowchart of step S40 in FIG. 5 in an embodiment.

FIG. 9 is a detailed flowchart of step S80 between step S40 and step S50 in an embodiment of the present invention.

1. Automatic walking equipment 2. Image acquisition device 3. Main control module

4. Walking module 5. Working module 6. Energy module

9. Driving wheel 10. Housing 11. Training wheel

12. Division unit 13. Color extraction unit 14. Calculation unit

15. Comparison unit 16. Storage unit 17. Texture extraction unit

18. Texture comparison unit 19. Recognition unit 20. Information extraction unit

21. Information changing unit 22. Judging unit 23. Sub-image block dividing unit

28. Target area 32. Correction unit 33. Recording unit

50. Working area 51. Non-working area 52 Island

detailed description

The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, so as to define the protection scope of the present invention more clearly.

FIG. 1 is a schematic diagram of an automatic walking device walking in a target area according to an embodiment of the present invention. The automatic walking device 1 can automatically walk on the ground or other working surfaces, and can also work while walking. The autonomous running device 1 can be an automatic vacuum cleaner, an automatic lawn mower, an automatic trimmer, and the like. In this embodiment, the self-propelled device is an automatic lawnmower. According to different working objects, the ground can be divided into a working area 50 and a non-working area 51 . The working area 50 refers to the area where the user wants the automatic walking device to walk through and perform work, and the non-working area 51 refers to the area that the user does not want the automatic walking device to pass through. In this embodiment, since the self-propelled device is an automatic lawn mower, its work is to perform mowing. Therefore, the walking area 50 can be but not limited to grass, and the non-working area 51 can be but not limited to concrete roads, big trees, ponds, fences, stakes, wall corners and the like. Usually, the grass is formed in pieces, and the non-walking area can be located around the grass, or can be surrounded by the grass to form an island 52, so the isolated island 52 is also a form of non-walking area. In the present invention, no boundary line may be set at the junction of the non-working area 51 and the working area 50 , and the automatic walking device 1 uses the visual difference between the working area 50 and the non-working area 51 for identification.

Referring to FIG. 2 and FIG. 3 , the self-propelled device 1 has a housing 10 and an image acquisition device 2 installed on the housing 10 . The image acquisition device 2 captures images of the area in front of the automatic walking device 1 . The ground area in front of the automatic traveling device 1 is the target area 28 for the automatic traveling device to walk. The target area 28 may be a working area, may also be a non-working area, or may be a collection of walking areas and non-walking areas. However, in order to be able to perform normal walking in the walking area, the automatic walking device 1 must identify the current target area 28 . Therefore, the self-propelled device 1 can take pictures of the target area 28 and form an image about the target area 28 by using the image acquisition device 2 . Therefore, the method for controlling the self-propelled device includes step S10 , that is, generating an image of the walking target area of the self-propelled device 1 . In this embodiment, the viewing range of the image acquisition device 2 is a fixed area, such as a fixed viewing angle ranging from 90 degrees to 120 degrees. In other optional embodiments, the viewing range can also be active, and a certain angle range within the viewing angle range can be selected as the actual viewing range, such as selecting a 90-degree range in the middle of the viewing angle range of 120 degrees as the actual viewing range. The image contains information of the target area, such as terrain relief, color distribution, and texture of the target area.

Please refer to FIG. 4 , in addition to the image acquisition device 2 , the automatic walking device 1 also includes a main control module 3 , a walking module 4 , a working module 5 and an energy module 6 . The main control module 3 is electrically connected to the walking module 4 , the working module 5 , the energy module 6 and the image acquisition device 2 , and plays a role in controlling the operation of the automatic walking device 1 .

The traveling module 4 includes a wheel set and a traveling motor for driving the wheel set. Wheel sets can be set up in a variety of ways. Usually, the wheel set includes a driving wheel 9 driven by a traveling motor and an auxiliary wheel 11 for auxiliary supporting the housing 10, and the number of the driving wheel 9 can be 1, 2 or more. As shown in FIG. 2 , the moving direction of the self-propelled device 1 is taken as the front side, the side opposite to the front side is the back side, and the two sides adjacent to the front and rear sides are the left and right sides respectively. In this embodiment, there are two driving wheels 9 of the self-propelled device 1 , which are respectively a left wheel 91 on the left side and a right wheel 92 on the right side. The left wheel 91 and the right wheel 92 are arranged symmetrically about the central axis of the self-propelled device 1 . The left wheel 91 and the right wheel 92 are preferably located at the rear of the housing 10, and the auxiliary wheels 11 are located at the front, and of course, in other embodiments, they can also be arranged alternatively.

In this embodiment, each of the left wheel 91 and the right wheel 92 is connected with a driving motor to realize a differential output to control the steering, so as to achieve the purpose of turning left or right. The left wheel 91 and the right wheel 92 can also output at constant speeds, so as to achieve the purpose of moving forward or backward. The drive motor can be directly connected to the drive wheel, but also a transmission can be set between the drive motor and the drive wheel 9, such as a common planetary gear train in the art. In other embodiments, two driving wheels and one driving motor may also be provided. In this case, the driving motor drives the left wheel 91 through the first transmission device, and drives the right wheel 92 through the second transmission device. That is, the same motor drives the left wheel 91 and the right wheel 92 through different transmission devices.

Work module 5 is used to perform specific work. In this embodiment, the working module 5 is specifically a cutting module, including a cutting unit (not shown) for mowing grass and a cutting motor (not shown) for driving the cutting unit.

The energy module 6 is used to provide energy for the operation of the automatic walking device 1 . The energy source of the energy module 6 can be gasoline, a battery pack, etc. In this embodiment, the energy module 6 includes a rechargeable battery pack set in the casing 2 . When working, the battery pack releases electric energy to maintain the automatic walking device 1 to work. During non-working hours, the battery can be connected to an external power source for supplemental power. In particular, for a more user-friendly design, when it detects that the battery is insufficient, the automatic walking device 1 will automatically search for a charging station (not shown) to supplement electric energy.

As shown in FIG. 3 , the image acquisition device 2 obtains the image of the target area 28 and transmits it to the main control module 3 . The main control module 3 includes a dividing unit 12 . The division unit 12 is used to divide the image into several cells. All the cells are combined into one image, and each cell occupies a part of the whole image. Each cell therefore contains identifying information for that part of the image. Each cell is roughly the same size. In addition, the several cells form a matrix array. The matrix array extends horizontally and vertically respectively. In the horizontal direction, about 20 cells are arranged in a row; in the vertical direction, about 20 cells are arranged in a row. In different embodiments, the number of cells arranged horizontally and vertically may be different. And each cell has at least one cell adjacent to it. For the cells located in the middle area of the array, each cell has 4 adjacent cells located in the top, bottom, left, and right respectively. In other words, the 4 cells are adjacent to that cell in the horizontal or vertical direction. . Of course, the meaning of adjacent is not limited to the four directions of up, down, left, and right. In another embodiment, the cell has 8 adjacent cells in eight directions: up, down, left, right, up left, up right, down left, and down right. In addition to being adjacent to the cell, it can also be adjacent to the cell in a direction that forms an angle of 45 degrees with the horizontal and vertical directions. And for the cells located in the edge area of the array, each cell may not have 4 adjacent cells, but there will be at least one adjacent cell. Therefore, the method for controlling the automatic walking device further includes step S20, that is, the image is divided into several cells, and each cell is adjacent to at least one other cell.

After dividing the cells, the division unit 12 starts to identify whether the target area corresponding to the cells is a walking area. The specific flow method is as follows: the main control module 3 first reads the identification information contained in each cell. In this embodiment, the identification information contained in the cell includes color information and texture information. In other embodiments, the information contained in the cells may be color information and other types of information. Since a cell is a part of an image, the image includes information about the target area. Therefore, the cell must contain the information of the corresponding target area, and of course the color information. Reading the identification information helps to determine whether the target area corresponding to the cell is a management working area or a non-working area. Because the grass as the working area is green, but the road and soil as the non-working area are not green, so if the color information of the cell is recognized as green, it can be considered that the cell corresponds to the walking area. If it is recognized that the color information is not green, it can be considered that the cell corresponds to a non-walking area. Of course, in order to further improve the accuracy, the non-walking area is also green in some cases. For example, some artificially processed objects are painted with green paint. In this case, the color of both the walking area and the non-walking area is green. From the color information It is not easy to distinguish the walking area from the non-walking area. Therefore, it is also necessary to add the identification of texture information. Because in the case that the non-walking area is also green, it usually has a regular texture, and although the grass in the walking area is also green, the growth of the grass is not so regular, so its texture is irregular. Furthermore, if it is recognized that the color information of the cell is green and the texture is irregular, it can be determined that the cell corresponds to a walking area. If the color is not green or the texture is regular, it can be determined that the cell corresponds to a non-walking area. Of course, in other embodiments, the purpose of distinguishing the walking area and the non-walking area can also be achieved by identifying other information, which will not be described in detail here.

For this, the main control module 3 also includes a color extraction unit 13 , a calculation unit 14 , a comparison unit 15 and a storage unit 16 . The main control module 3 extracts the color information of the cell, then compares the color information with the preset information, and identifies whether the cell is a walking area according to the comparison result. The specific method is as follows: Since each cell actually contains many pixel units, the color displayed by the pixel unit is unique. Therefore, the function of the color extracting unit 13 is to extract the color of each pixel unit in the cell, in particular, three primary colors (RGB) components are extracted. The preset information refers to preset information that functions as a reference comparison object. In this embodiment, the preset information refers to the numerical ranges of the three primary color components of the predetermined color. In this embodiment, the predetermined color refers to green. Comparing the three primary color components of a pixel with the three primary color components of a predetermined color, if the three primary color components of a pixel respectively fall within the value range of the three primary color components of the predetermined color, then it is judged that the color of the pixel is the predetermined color. Note that if it does not fall within the value range, the color of the pixel is judged to be an unpredetermined color. In another embodiment, the storage unit 16 has a preset hue value (Hue) range of a predetermined color, and after extracting the three primary color components of a pixel, the RGB components obtained are further converted into HSV (hue Hue, saturation, brightness Value ) value, and compare whether its tone value is within the preset tone value range, if yes, it is determined that the color of the pixel is a predetermined color, otherwise, it is determined that the color of this pixel is a non-predetermined color.

Then the calculation unit 14 calculates the ratio of the number of pixels with a predetermined color to the total number of pixels in a cell (hereinafter referred to as proportion). The comparison unit 15 then compares the ratio with a first preset value, and if the ratio exceeds or reaches the first preset value, it is determined that the color display of the cell is a predetermined color. The first preset value can be 50%, 60% or other values. In addition, the first preset value can be stored in the storage unit 16 .

Combined with some other information of the cell, it can be identified that the cell belongs to the working area or the non-working area. In this embodiment, it refers to the texture information of the cell. The main control module 3 also includes a texture extraction unit 17 and a texture comparison unit 18 . The texture extracting unit 17 extracts the texture feature value of the cell. The degree of dispersion of at least one parameter of all pixels in the cell may reflect the degree of difference between values of the parameter. If the target area is green paint, the dispersion of a parameter in the image is very small, even 0. Due to the irregular texture of the grass, the dispersion of a parameter difference value of all pixels of the cell will be greater than or equal to a preset dispersion, thus reflecting the irregularity of the texture of the cell. Therefore, in this embodiment, the texture feature value is parameter dispersion, such as color dispersion, grayscale dispersion, brightness dispersion, and the like.

The texture comparison unit 18 compares the texture feature value of the cell with a second preset value to determine whether the texture feature value reaches the second preset value. In this embodiment, the second preset value is a preset dispersion. The texture comparison unit 18 can exist independently, or can be integrated into the comparison unit 15 . The second preset value can also be pre-stored in the storage unit 16 .

The main control module 3 also includes an identification unit 19 . The color extraction unit 13 , the calculation unit 14 , the comparison unit 15 and the storage unit 16 may constitute a part of the recognition unit 19 in one embodiment, or be integrated into the recognition unit 19 to form a whole. It can also be used as a unit component juxtaposed with the recognition unit 19 in another embodiment. When the recognition unit 19 recognizes that the proportion of pixels with a predetermined color in the cell reaches or exceeds the first preset value and the texture feature value of the cell reaches or exceeds the second preset value, it determines the target corresponding to the cell The area is a walking area; if the ratio does not reach the first preset value or the texture feature value does not reach the second preset value, it is determined that the target area corresponding to the cell is a non-walking area. Therefore, the method for controlling the automatic walking device further includes step S30, that is, reading the identification information contained in each cell and identifying it, so as to obtain the identification result of whether the target area corresponding to the cell is a working area.

The recognition unit 19 of the main control module 3 respectively recognizes all cells in the image, so as to obtain recognition results of all cells. In a preferred embodiment, the main control module 3 further includes a correction unit 32, and the correction unit 32 corrects the identification result of the cell based on the Markov stochastic model. Therefore, in this embodiment, the control method further includes step S60, that is, correcting the different recognition results in the cells based on the smoothing process. This is because in actual working conditions, the recognition result obtained through step S30 may have certain errors, that is, an abnormal recognition result will be generated. The correction process can correct the abnormal recognition result, thereby improving the recognition accuracy. Specifically, for each cell in the image, there must be adjacent cells. The purpose of correction can be achieved by comprehensively considering the recognition results of its adjacent cells and the recognition results of the cell itself. The modification unit 32 includes the information extraction unit 20 and the information modification unit 21 .

In one embodiment, the correction method is as follows: step S60 includes steps S61, S62, S63 and S64. Wherein step S61 means that for each cell, the information extraction unit 20 extracts the recognition results of all adjacent cells with the cell; The number of adjacent cells, and the proportion of this number to the total number of adjacent cells. For example, if the recognition result of the cell is a working area, the calculation unit 14 counts the number of the recognition results in its adjacent cells that are also working areas, and calculates that the number accounts for the total number of adjacent cells. proportion. Step S63 means that the comparison unit 15 compares the ratio with the fourth preset value, if the ratio is greater than or equal to a fourth preset value (usually the fourth preset value is not less than 50%, it can be is 50%, 75%, etc.), indicating that the adjacent cells of the recognition result occupy the majority of all adjacent cells, so the cell information change unit 21 keeps the recognition result of this cell unchanged. If the proportion is less than the fourth preset value, the cell information changing unit 21 changes the recognition result of the cell to another recognition result, for example, the original recognition result of the cell is the working area, and the recognition result is becomes a non-working area. In the whole process, for example, for a cell, the original recognition result is a non-working area. However, the recognition results of 3 cells in the 4 adjacent cells are working areas, and this ratio (3/4=75%) is greater than the fourth preset value (assuming that the fourth preset value is 50%) , then according to the result of the adjacent cell, it is determined that the original recognition result of this cell is the same as the recognition result of the adjacent cell, so the recognition result of this cell remains unchanged, and it is still the working area. If the recognition result of only one of the 4 adjacent cells is the working area, and the ratio (1/4=25%) is less than the fourth preset value, then the recognition result of the cell is determined to be adjacent to it The recognition result of the cell is inconsistent, and the recognition result of the cell may be caused by an error, so the recognition result of the cell is corrected as the working area. Of course, it is emphasized again that the adjacency here is not limited to the adjacency in the four directions of up, down, left, and right, and may not be limited to a total of eight directions such as up-left, up-right, down-left, and down-right. Similarly, the original identification result of the cell is not limited to the working area, and may also be the non-working area. The final step S64 is to apply this method to all cells, that is, complete the result correction of the entire image, that is to say, perform the above steps S61-S63 on all cells to correct the recognition results of all cells.

In another embodiment, step S60 includes steps S66, S67, S68, S69, S70, S71 and S72. Step S66 first obtains the reliability of the recognition result of the cell. Reliability is usually a value between 0 and 100%. Certainly, the reliability may also be a numerical value in other forms. In step S67, the degree of reliability is marked as Y1, and the degree of unreliability is marked as N1, where N1=1-Y1. Y1 can also be called similar reliability. N1 can be called dissimilar reliability. Y1 and N1 may be stored in the storage unit 16 . Then step S68 is used to obtain and calculate the reliability of the cells adjacent to the cell. Reliability is usually a value between 0 and 100%. If the cell has 8 adjacent cells, obtain 8 similar reliability and dissimilar reliability through step S69 and step S67 similarly. Specifically, the reliabilities of the eight adjacent cells are denoted as similar reliabilities Ya, Yb, Yc . . . , and the dissimilar reliabilities are denoted as Na, Nb, Nc . . . Then, through step S70, the eight similar reliability levels are weighted and summed to obtain Y2. In this embodiment, the weight coefficients are consistent in size, preferably 1/8. Of course, the weight coefficients may also take different values. Similarly, the eight dissimilar reliability levels are weighted and summed to obtain N2. The weight coefficients of the eight dissimilar reliability degrees may be the same, and may be consistent with the weight coefficients of the similar reliability degrees. Then, through step S71, compare the sizes of Y1+αN1 and Y2+αN2 and take corresponding measures. Specifically, α here is a weight coefficient, and it may be different from or the same as the weight value of the previous steps. Further, the comparison process can be performed in the comparison unit 15 or in other components. Regarding the result of the comparison, if the result of Y1+αN1 is greater than or equal to the result of Y2+αN2, the information changing unit 21 maintains the identification result of the cell unchanged, if the result of Y1+αN1 is less than the result of Y2+αN2, the information The changing unit 21 changes the recognition result of the cell. Then, through step S72, the above-mentioned process is performed on all the cells in the image, and each cell will participate in an iterative cycle until the recognition results of all cells do not change any more.

Therefore, the method for controlling the automatic walking device further includes step S40. Step S40 is used to determine whether the target area corresponding to the sub-image block including several cells is a walking area. The self-propelled device 1 includes a judging unit 22 for performing this step. The judging unit 22 includes a sub-image block dividing unit 23, and the sub-image block dividing unit 23 is used to divide the image into several sub-image blocks. In one of the embodiments, the specific manner of division is as follows: step S40 includes steps S41, S42, and S43. First, the sub-image block dividing unit 23 selectively divides the image into several sub-image blocks according to the walking direction of the automatic walking device through step S41. Each sub-image block corresponds to a different walking direction. In one of the embodiments, the sub-image block dividing unit 23 divides the image into three sub-image blocks in the middle, left, and right, corresponding to the sub-areas in the target area, respectively. As shown in Figure 3, the middle part corresponds to the middle area a in the middle of the front of the self-propelled device 1 and is as wide as the self-propelled device 1; Area b; the right part corresponds to the front of the automatic walking device 1 and the right area c on the right side of the middle area a. Each of the three sub-image blocks contains a plurality of cells. In another embodiment, the sub-image block division unit 23 may further divide the image into five different sub-image blocks, namely front, left front, left, right front, and right. Since each sub-image block contains several cells, the judging unit 22 judges whether the target area corresponding to the sub-image block is a walking area or a non-walking area based on the identification results of all the cells in the sub-image block. Specifically, it is assumed that a total of 60 cells in three rows of cells located at the front end of the image are used as the middle sub-image block. In this embodiment, the information extraction unit 20 of the automatic walking device 1 extracts the recognition results of all cells in the middle sub-image block, and the calculation unit 14 calculates the number of cells whose recognition result is the working area area, and marks the number for A. Of course, in other embodiments, the number of cells identified as non-working areas can also be counted. The comparison unit 15 compares the number of cells whose recognition result is the walking area with the size of the third preset value. When the quantity A or the ratio of A to all cells in the sub-image block is greater than or equal to a third preset value, the judging unit 22 may determine that the sub-image block is a walking area. Of course, it can also be set such that when the number of cells in the non-walking area is less than a third preset value as a recognition result, it can be determined that the sub-image block is a walking area. The third preset value in this embodiment is pre-stored in the storage unit 16 and may be 30, 40, 50 and other values. In other embodiments, the automatic walking device 1 can also use the ratio of the recognition result that the cells in the walking area or non-walking area to all the cells in the sub-image block as a parameter, and compare it with another third preset value , the third preset value in this embodiment is greater than or equal to 50%, and may be 50%, 60%, 90% and so on.

After judging that the target area corresponding to the sub-image block is a walking area or a non-walking area, the automatic walking device 1 controls the automatic walking device to move forward, backward, turn left or turn right according to the judgment result through step S50. According to the judging result of the judging unit 22, the self-propelled device 1 will execute a specific response action. The actions that the walking module 4 controls the automatic walking device 1 to respond to include: forward (F), backward (B), left turn (L), right turn (R) and no change (N). In the embodiment in which the image is divided into three sub-image blocks, left, middle, and right, each sub-image block has a walking area and a non-walking area respectively in the recognition result. So there are a total of eight different situations: 1. The left, middle and right are walking areas; 2. The left middle is the walking area, and the right is the non-walking area; 3. The left and right are walking areas, and the middle is the non-walking area; 4. Left 5. The left is the non-walking area, and the middle and right is the walking area; 6. The left and right are non-walking areas, and the middle is the walking area; 7. The middle left is the non-walking area, and the right is the walking area 8. The left, middle and right are non-walking areas.

In the first case, the main control module 3 makes the walking module 4 perform an action that does not change (N);

In the second case, the main control module 3 makes the walking module 4 perform the action of turning left and moving forward (LF);

In the third case, the main control module 3 makes the walking module 4 perform the action of turning backward and turning left and moving forward (BLF);

In the fourth case, the main control module 3 makes the walking module 4 perform the action of turning backwards and turning left and moving forward (BLF);

In the fifth case, the main control module 3 makes the walking module 4 perform the action of turning right and advancing (RF);

In the 6th case, the main control module 3 makes the walking module 4 perform the action of going back and turning right and moving forward (BRF);

In the 7th case, the main control module 3 makes the walking module 4 perform the action of retreating and turning right and moving forward (BRF);

In the eighth case, the main control module 3 makes the walking module 4 perform an action of turning backward and turning right and forward (BRF) or turning backward and turning left and forward (BLF).

Further illustrate, when judging that the current target area is a walking area, the automatic walking device 1 may continue to execute the original walking strategy, such as maintaining the original walking state; Then the walking direction will be changed, and further, the walking direction can be selectively walked away from the sub-image block. Since there are multiple sub-image blocks in the image, the automatic walking device 1 needs to identify the walking area or the non-walking area for the multiple sub-image blocks, and then adopt corresponding strategies. In a preferred embodiment, the automatic walking device can identify the multiple sub-image blocks at the same time. For example, for the three sub-image blocks such as the middle part, the left part, and the right part, if it is detected that the three sub-image blocks are all walking areas, the automatic walking device keeps moving forward; if the three sub-image blocks are detected The image blocks are all non-walking areas, and the automatic walking equipment will turn 180 degrees and move backward; if it detects that the sub-image blocks in the middle and left are all non-walking areas, and the right part is a walking area, the automatic walking equipment will move towards The direction away from the middle and the left is to move to the lower right. Of course, you can go back and then turn right, or turn right first and then go back.

Of course, in a preferred embodiment, the sub-image block division unit 23 of the automatic walking device 1 may also include a process of dividing the sub-image blocks multiple times and then performing comprehensive judgment. The regions corresponding to the sub-image blocks divided each time may be different. In this way, the judgment results of different areas are comprehensively considered, avoiding errors in strategy formulation caused by inaccurate judgment results in a single area, and improving the walking accuracy of the automatic walking device 1 . Specifically still take the three rows of cells located at the front end of the image as an example of the middle sub-image block. In one judgment, the object of the sub-image block is the 60 cells. In another recognition process, it can be A total of 80 cells in the four rows of cells at the front end of the image are used as the middle sub-image block, and the object of the sub-image block is judged to be the 80 cells. In addition, the third preset values used in the two judgments are also different, here may be but not limited to 60. Combine these two recognitions as a new basis for judgment. For example, when the judgment condition is that 40 of the 60 cells forming three rows are identified as walking areas, and 60 of the 80 cells forming four rows When a cell is identified as a walking area, the middle part can be identified as a walking area. If the two judging conditions cannot be satisfied at the same time, the middle part is determined to be a non-walking area. Of course, the same way can also be used to judge the left part and the right part.

In another preferred embodiment, because a single image may still have distortions in the information collection of the target area, for example, objects that pass quickly at a certain moment will cast shadows on the target area, thereby affecting the automatic The process of judging the target area by the walking equipment. Therefore, in this embodiment, the self-propelled device further includes a step S80 between the above step S40 and step S50. Through step S80, comprehensive filtering is performed according to the sub-image blocks in the multiple images to obtain the final judgment result of whether the sub-image blocks are walking areas. The target area can be photographed multiple times within a certain period of time, thereby forming a multi-frame image. Then, the judgment information contained in each frame of image is comprehensively filtered to obtain the final judgment result.

Method step S80 includes at least steps S81, S82 and step S84. In a preferred embodiment, step S80 further includes step S83 between S82 and S84. The specific method is as follows: through step S81, the image acquisition device 2 shoots the same target area multiple times to form multiple frames of images, and each frame of images is respectively referred to as the first frame image, the second frame image...the Nth frame image. The self-propelled device 1 further includes a recording unit 33, and the recording unit 33 is used to process the weight value according to the judgment result of the sub-image block through step S82. Specifically, when the judging unit 22 judges that the first frame of image is a walking area, the recording unit 33 will add a fifth preset value to the initial weight value. For the convenience of description, the initial weight value can be marked as 0, and of course it can also be marked as other values. The fifth preset value may be a preset fixed constant, which is also a function of variation. In this embodiment, the fifth preset value may be but not limited to 3. When the recognition result of the first frame image walks in the area, the recording unit changes the corresponding weight value to 3. Then the recognition result of the second frame image is processed. If the recognition result of the second frame of image is also a walking area, the recording unit 33 of the automatic walking device 1 adds a fifth preset value to the current weight value. At this time, the corresponding weight value becomes 6. If the recognition result of the second frame of image is not a walking area, the recording unit 33 does not change the current weight value. Then, the recognition result of the third frame image is processed. If the recognition result of the third frame image is also a walking area, the current weight value becomes 9. And so on until the Nth frame image. In addition, through step S84, the comparison unit 15 also compares the current weight value with a seventh preset value. When the current weight value is greater than or equal to the seventh preset value, it is determined that the judgment result is correct, that is, the current target area is indeed a walking area. The seventh preset value can be set to 8, for example. In this way, the recognition results of multiple frames of images are comprehensively considered, thereby avoiding adverse effects caused by possible erroneous results of a single frame of images. Further, the identification of each frame of image can be decomposed into the identification of each sub-image block of the image. For example, each frame of image can be decomposed into three sub-image blocks, left, middle, and right, and the recording unit can record the three sub-image blocks respectively, corresponding to three weight values.

In addition, in a preferred embodiment, step S83 also includes that during the frame-to-frame image switching process, the recording unit 33 also subtracts a sixth preset value from the current weight value, so that the current weight value reaches or exceeds the sixth preset value. The seven presets are worth a longer time, allowing more frames of images to be considered comprehensively, further improving accuracy. In this embodiment, the sixth preset value may be, but not limited to, 1. For example, the first frame image recognition result is a walking area, and the current weight value becomes 3. When the recognition result of the second frame image is a non-walking area, the current weight value becomes 2. When the recognition result of the third frame image is the walking area, the current weight value becomes 4. When the recognition result of the fourth frame image is the walking area, the current weight value becomes 6. In this way, until the weight value reaches or exceeds the seventh preset value, the current target area is determined to be the walking area. If the weight value has not reached the seventh preset value, it is determined that the current target area is a non-walking area. Of course, those skilled in the art can imagine that through some changes, when the weight value reaches or exceeds the seventh preset value, it is determined that the current target area is a non-walking area, and when the weight value has not reached the seventh preset value, The current target area is identified as the walking area. That is, exchange the recognition conditions of the walking area and the non-walking area. Moreover, the calculation rules for the weight value are further refined. For example, it can be set that if the weight value is reduced to the minimum value under any conditions, for example, when the weight value is reduced to 0, it will not continue to decrease.

The present invention is not limited to the structures of the specific embodiments cited, and the structures based on the concept of the present invention all belong to the protection scope of the present invention.

Claims (8)

1. a kind of identification method about the target zone of self-propelled equipment walking, it is characterized in that, described identification method comprises the following steps: S10. Acquiring an image of the walking target area of the automatic walking device; S20. Divide the image into several cells, each cell has at least one adjacent cell; S30. Identify whether the target area corresponding to the cell is a working area according to the color information of the specified pixel in the cell and the texture feature value of the cell, and obtain an identification result; S60. For each cell, change or maintain the recognition result obtained in step S30 according to the recognition result of adjacent cells. 2. The identification method according to claim 1, characterized in that: said S60 step further comprises the following steps: S61. Designate a cell and obtain its recognition result; S62, counting the number of adjacent cells having the same recognition result as in step S61; S63, calculating the ratio of the quantity in step S62 to the total number of adjacent cells; S64. If the ratio exceeds or reaches the fourth preset value, then maintain the recognition result of the cell specified in step S61 unchanged; if the ratio is smaller than the fourth preset value, then change the recognition result of the cell specified in step S61, Wherein the fourth preset value is greater than or equal to 50%; S65. Execute the above steps S61-S64 for all the cells. 3. The identification method according to claim 2, wherein the adjacent cells include cells adjacent to the cell horizontally and vertically. 4. The identification method according to claim 3, wherein the adjacent cells further include cells adjacent to the cells in a direction forming an angle of 45 degrees with the horizontal direction and the vertical direction. 5. The identification method according to claim 1, characterized in that: said S60 step further comprises the following steps: S66. Designate a cell, and acquire the reliability Y1 of the cell for its recognition result, where the reliability Y1 is a value between 0% and 100%; S67. Calculate 1-Y1, and mark the result as N1; S68. Obtain the reliability Ya, Yb... of all adjacent cells of the specified cell for the recognition result thereof, where the reliability Ya, Yb... is a value between 0% and 100%; S69, calculate 1-Ya, 1-Yb..., and mark the result as Na, Nb...; S70. The weighted sum of Ya, Yb... to obtain the weighted sum Y2, and the weighted sum of Nb, Nc... to obtain the weighted sum N2, wherein the weighted coefficients here are all the same; S71. Calculate and compare the results of Y1+αN1 and Y2+αN2 respectively, where α is a coefficient; If Y1+αN1 is greater than or equal to Y2+αN2, then maintain the recognition result of the specified cell unchanged, if the result of Y1+αN1 is less than the result of Y2+αN2, then change the recognition result of the specified cell; S72. Perform the above steps S66-S71 for all the cells until the identification results of all the cells do not change any more. 6. An automatic walking device, characterized in that: comprising a housing, an image acquisition device positioned on the housing, the image acquisition device is used to photograph the target area and generate an image, and drives the walking module of the automatic walking device to walk, Connect the image acquisition device and the walking module to control the main control module of the automatic walking equipment, wherein the main control module divides the unit, the identification unit and the correction unit, and the division unit divides the image into several cells , the identification unit identifies whether the target area corresponding to the cell is a working area, and transmits the identification result to the correction unit, and the correction unit changes or maintains each cell according to the identification results of adjacent cells The recognition result obtained by the recognition unit. 7. The self-propelled device according to claim 6, wherein the cells adjacent to the cell include cells adjacent to the cell horizontally and vertically. 8. The self-propelled device according to claim 7, characterized in that: the cells adjacent to the cell also include a direction adjacent to the cell in a direction at an angle of 45 degrees to the horizontal and vertical directions. of cells.