A method for recognizing and analyzing clothes of a washing and care robot based on image processing technology

Abstract

The present invention belongs to the technical field of clothing recognition, and discloses a method for clothing recognition and analysis of a washing and care robot based on image processing technology. When the present invention conducts clothing recognition, it analyzes the possibility of color fading of a specified clothing based on the collected image, and then classifies the specified clothing. This analysis method can intuitively and accurately understand the actual condition of the clothing and avoid the limitations of material label information. The image-based analysis method is not limited by the type, material or label of the clothing, and can be applied to various different types of clothing, improving the processing efficiency. When the present invention analyzes the possibility of color fading of a specified clothing, it comprehensively considers the proportion of the white area and the color fading possibility evaluation index of the non-white area (monitoring area), and then classifies each specified clothing into clothing with the possibility of color fading and clothing without the possibility of color fading, and performs cleaning operations for different types of specified clothing respectively, which can more precisely evaluate the overall color fading risk of the clothing.

Description

Technical Field

[0001] The present invention belongs to the technical field of clothing recognition, and relates to a method for recognizing and analyzing clothing of a washing and care robot based on image processing technology. Background Art

[0002] In the operation process of a washing and care robot, clothing recognition is the crucial first step. Accurate clothing recognition enables the robot to select appropriate washing modes, detergent dosages, and washing times according to the characteristics of different clothes, such as color, stain degree, etc., so as to achieve efficient, accurate, and safe clothing washing and care. Therefore, the research on clothing recognition and analysis of a washing and care robot based on image processing technology is of great significance.

[0003] In the prior art, there are also related solutions for clothing recognition technology. For example, a Chinese patent application for an intelligent washing and care method, device, system, equipment, and readable storage medium with the publication number CN117935277A includes: obtaining target image data corresponding to a target item group through a client. Based on the target image data, obtaining item attribute information corresponding to the target item group through a processing end. The item attribute information includes target color information, target material information, and target stain information. Based on the target material information, determining a recommended washing mode corresponding to the target item group through the processing end, and based on the target color information and target stain information, determining target washing parameters corresponding to the target item group. Washing the target item group based on the recommended washing mode and target washing parameters through a device end.

[0004] In addition, a Chinese patent application for a method and device for recognizing clothing materials with the publication number CN113362269A includes: obtaining image information of a clothing item to be processed. Performing image recognition on the image information according to a material recognition algorithm to obtain the actual material information of the clothing item to be processed. Comparing the actual material information with preset standard material information corresponding to the clothing item to be processed to obtain a material recognition result of the clothing item to be processed. Pushing the material recognition result of the clothing item to be processed to a user. The embodiments of the present invention can achieve the effects of reducing the hardware cost of distinguishing the authenticity of clothing materials and simplifying the operation and saving time.

[0005] Although the above two solutions propose some solutions for clothing recognition, there are still certain limitations: on the one hand, existing technical solutions all identify the tags of clothing to determine data such as the material of the clothing. This analysis method cannot be applied to the recognition of all clothing, reducing the applicability of the system and affecting the clothing recognition efficiency of the system. On the other hand, after the existing clothing recognition, the analysis of the clothing washing process uses the cleaning parameters corresponding to the clothing with the highest priority as the current cleaning parameters. This way of setting parameters lacks a refined classification analysis of clothing washing, which may cause over-washing of some clothing, affect the washing effect, cause waste of resources and the accuracy of system analysis. Summary of the Invention

[0006] In view of this, to solve the problems raised in the above background technology, a clothing recognition and analysis method for a washing and care robot based on image processing technology is proposed.

[0007] The object of the present invention can be achieved by the following technical solutions: A clothing recognition and analysis method for a washing and care robot based on image processing technology, including: S1. Designated clothing image acquisition: Use a high-definition camera to collect images of each designated clothing, and the images specifically include front images and back images.

[0008] S2. White area ratio analysis: Analyze the white area ratio of each designated clothing based on the images of each designated clothing.

[0009] S3. Color fading possibility analysis of designated clothing: Determine the monitoring areas of each designated clothing based on the images of each designated clothing, analyze the color consistency and shape irregularity evaluation of each monitoring area of each designated clothing corresponding to adjacent non-white areas, and then analyze the color fading possibility of each designated clothing.

[0010] S4. Designated clothing classification: Classify each designated clothing based on the white area ratio and color fading possibility of each designated clothing to obtain the types of each designated clothing. The types of each designated clothing specifically include clothing with a color fading possibility and clothing without a color fading possibility, and then clean them separately.

[0011] S5. Stain situation analysis: Locate each stain area of each designated clothing based on the images of each designated clothing, and analyze the stain evaluation of each stain area of each designated clothing. The stain evaluation specifically includes stain color analysis and stain texture analysis.

[0012] S6. Stain type identification: Identify the stain types of each stain area of each designated clothing based on the stain evaluation of each stain area of each designated clothing, and then select corresponding cleaning methods and cleaning agents for different types of stains, and then determine the cleaning methods and cleaning agents for different types of designated clothing.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) When the present invention performs clothing recognition, it analyzes the possibility of color fading of the specified clothing based on the collected images, and then classifies the specified clothing. This analysis method can intuitively and accurately understand the actual condition of the clothing, avoiding the limitations of material label information. The image-based analysis method is not limited by clothing types, materials or labels, and can be applied to various different types of clothing, including clothing without labels or with incomplete label information, enabling fast and accurate clothing recognition and classification, and greatly improving the processing efficiency.

[0014] (2) When the present invention analyzes the possibility of color fading of the specified clothing, it comprehensively considers the proportion of the white area and the color fading possibility evaluation index of the non-white area (monitoring area), and then classifies each specified clothing into clothing with the possibility of color fading and clothing without the possibility of color fading, and performs cleaning operations on different types of specified clothing respectively. This analysis method can more finely evaluate the overall color fading risk of the clothing. For different types of clothing, different cleaning strategies can be adopted to reduce the staining risk, which helps to protect the clothing and extend its service life.

[0015] (3) When the present invention performs cleaning of the specified clothing, it identifies the stain type based on the analysis results of the stain color and stain texture of each stain area of each specified clothing. This analysis method accurately identifies the stain type through the analysis of stain color and texture. The washing and care robot can select the most suitable detergent and cleaning parameters for each type of stain, thereby improving the cleaning effect. Precise stain type identification enables the washing and care robot to more effectively remove stains during a single cleaning process. Description of the Drawings

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0017] Figure 1 It is a schematic diagram of the implementation steps of the method of the present invention.

[0018] Figure 2 It is a flowchart of an embodiment for classifying specified clothing provided by the present invention.

[0019] Figure 3 It is a schematic diagram of an embodiment for identifying stain types provided by the present invention. Detailed Embodiments

[0020] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] Please refer to Figure 1 As shown, the present invention provides a method for identifying and analyzing clothes of a washing and care robot based on image processing technology, including: S1. Specified clothing image acquisition: Use a high-definition camera to collect images of each specified clothing, and the images specifically include front images and back images.

[0022] It should be noted that during the image acquisition process, it is necessary to ensure that the specified clothing is as flat as possible and unobstructed.

[0023] S2. Analysis of the proportion of white areas: Analyze the proportion of white areas of each specified clothing based on the images of each specified clothing.

[0024] In a preferred embodiment of the present invention, the specific analysis method for analyzing the proportion of white areas of each specified clothing is as follows: Extract the front images and back images of each specified clothing, and then use image processing software to obtain the white area area of the front image, the white area area of the back image, the clothing area of the front image, and the clothing area of the back image corresponding to each specified clothing.

[0025] Calculate the proportion of the white area of the front image of each specified clothing to the clothing area of the front image to obtain the proportion of the white area of the front image of each specified clothing, and calculate the proportion of the white area of the back image of each specified clothing to the clothing area of the back image to obtain the proportion of the white area of the back image of each specified clothing.

[0026] Calculate the average value of the proportion of the white area of the front image and the proportion of the white area of the back image of each specified clothing to obtain the proportion of the monitored white area of each specified clothing.

[0027] S3. Analysis of the possibility of color fading of specified clothing: Determine the monitored areas of each specified clothing based on the images of each specified clothing, analyze the color consistency and shape irregularity evaluation of each monitored area corresponding to adjacent non-white areas of each specified clothing, and then analyze the possibility of color fading of each specified clothing.

[0028] In a preferred embodiment of the present invention, the analysis of the color consistency situation requires constructing a color consistency index for each specified clothing monitoring area corresponding to each adjacent non-white area. The specific method is as follows: Extract the front and back images of each specified clothing, and then use image processing software to obtain the area of each non-white area of each specified clothing. Compare the areas of the non-white areas of each specified clothing, and select the non-white area corresponding to the largest area as the monitoring area.

[0029] It should be explained that when analyzing the color consistency index of each specified clothing monitoring area corresponding to each adjacent non-white area, the reason for selecting the non-white area corresponding to the largest area as the monitoring area for analysis is as follows: 1. It can more representatively reflect the overall color characteristics of the non-white part of the clothing. Since the color of a large area is usually more stable and less affected by local color differences, using it as the analysis object can more accurately evaluate the color consistency of the clothing. 2. In actual operation, analyzing the color consistency index of multiple small areas may consume a lot of time and effort. By selecting the area with the largest area for analysis, while ensuring the accuracy of the analysis, the analysis efficiency can be improved and the operation cost can be reduced.

[0030] Use image processing software to obtain the chromaticity values of each specified clothing monitoring area and the chromaticity values of the corresponding adjacent non-white areas.

[0031] Compare the chromaticity values of each specified clothing monitoring area and the chromaticity values of the corresponding adjacent non-white areas, and analyze to obtain the color consistency index of each specified clothing monitoring area and each adjacent non-white area.

[0032] It should be supplemented that the specific method for analyzing the color consistency index of each specified clothing monitoring area and each adjacent non-white area is as follows: If the chromaticity value of a specified clothing monitoring area is the same as the chromaticity value of a corresponding adjacent non-white area, then set the color consistency index of the specified clothing monitoring area and the adjacent non-white area to 1.

[0033] If the chromaticity value of a specified clothing monitoring area is different from the chromaticity value of a corresponding adjacent non-white area, calculate the absolute value of the difference between the chromaticity value of the specified clothing monitoring area and the chromaticity value of the corresponding adjacent non-white area to obtain the chromaticity value deviation amount between the specified clothing monitoring area and the adjacent non-white area. Then, calculate the ratio of the preset reference chromaticity value deviation amount to the chromaticity value deviation amount between the specified clothing monitoring area and the adjacent non-white area respectively to obtain the color consistency index of the specified clothing monitoring area and the adjacent non-white area.

[0034] Furthermore, obtain the color consistency index of each specified clothing monitoring area and each adjacent non-white area.

[0035] It should be explained that the color consistency index between each specified clothing monitoring area and each adjacent non - white area reflects the color consistency between the color of each specified clothing monitoring area and the color of the corresponding adjacent non - white area. In the chromaticity value analysis method, the colors corresponding to adjacent chromaticity values are similar, and the smaller the deviation of the chromaticity value, the closer the corresponding colors are. In practice, if the clothing fades, it may cause a change in the color of the color area. Therefore, the closer the chromaticity values of adjacent color areas are, the higher the possibility of fading. Combining other influencing indicators (such as shape irregularity) to comprehensively judge whether the corresponding color area is a faded area. If the chromaticity value of a monitoring area differs greatly from the chromaticity value of the corresponding adjacent non - white area, it reflects a large color difference between the two areas (such as red, blue), and in this case, the possibility of fading is relatively low.

[0036] In a preferred embodiment of the present invention, the analysis of the shape irregularity evaluation situation requires constructing a shape irregularity evaluation index for each specified clothing monitoring area corresponding to each adjacent non - white area, and the specific method is as follows: Using image - processing software to obtain the area and perimeter of each specified clothing monitoring area corresponding to each adjacent non - white area, and respectively denote them as 、 where represents the number of the specified clothing, , represents the quantity of the specified clothing, represents the number of the adjacent non - white area, , represents the quantity of the adjacent non - white area.

[0037] Using the formula to analyze and obtain the shape irregularity evaluation index for each specified clothing monitoring area corresponding to each adjacent non - white area, where represents a pre - set reference area, represents a pre - set reference perimeter.

[0038] It should be explained that the reason for selecting the area and perimeter of each specified clothing monitoring area corresponding to each adjacent non - white area as the shape irregularity evaluation index: The shape irregularity evaluation index is a quantitative index used to measure the degree of deviation of an object's shape from a regular shape (such as a circle, square, etc.). It can help us more objectively describe the complexity and irregularity of the object's shape. In two - dimensional geometric analysis, there is a certain relationship between the regularity of a graph and its perimeter and area. Under the same area condition, the longer the perimeter, the more irregular the corresponding image.

[0039] In a preferred embodiment of the present invention, to analyze the color fading possibility of each specified piece of clothing, it is necessary to construct an evaluation index for the color fading possibility of the monitored area of each specified piece of clothing. The specific method is as follows: Extract the color consistency index and the shape irregularity evaluation index of the monitored area of each specified piece of clothing and each adjacent non-white area, and then calculate the sum according to the weights to obtain the evaluation index for the color fading possibility of the monitored area of each specified piece of clothing.

[0040] Exemplarily, the weights corresponding to the color consistency index and the shape irregularity evaluation index of the monitored area of each specified piece of clothing and each adjacent non-white area are 0.6 and 0.4.

[0041] It should be explained that the setting basis for the weights corresponding to the color consistency index and the shape irregularity evaluation index of the monitored area of each specified piece of clothing and each adjacent non-white area: Based on the requirements for stain recognition accuracy, color space characteristics and visual perception factors, the association between washing procedures and stain types, as well as the importance of stain shape characteristics, the combination with clothing styles and structures, cleaning difficulty and targeted cleaning strategies, etc. The comprehensive consideration aims to accurately locate stains, distinguish stain types and determine appropriate cleaning methods to improve the cleaning efficiency and quality of the washing and care robot.

[0042] S4. Classification of specified clothing: Classify each specified piece of clothing based on the proportion of the white area and the color fading possibility of each specified piece of clothing to obtain the type of each specified piece of clothing. The types of each specified piece of clothing specifically include clothing with the possibility of color fading and clothing without the possibility of color fading, and then clean them separately.

[0043] In a preferred embodiment of the present invention, the specific method for classifying each specified piece of clothing is as follows: Please refer to Figure 2 As shown, extract the proportion of the monitored white area of each specified piece of clothing and the evaluation index for the color fading possibility of the monitored area, and then compare them with the pre-set threshold for the proportion of the monitored white area and the threshold for the evaluation index for the color fading possibility of the monitored area respectively.

[0044] Exemplarily, the threshold for the proportion of the monitored white area is 0.8, and the threshold for the evaluation index for the color fading possibility of the monitored area is 0.85.

[0045] If the proportion of the monitored white area of a specified piece of clothing is less than the threshold for the proportion of the monitored white area and the evaluation index for the color fading possibility of the monitored area is greater than the threshold for the evaluation index for the color fading possibility of the monitored area, it is determined that the specified piece of clothing is clothing with the possibility of color fading. Otherwise, it is determined that the specified piece of clothing is clothing without the possibility of color fading.

[0046] When analyzing the possibility of color fading of specified clothing, the present invention comprehensively considers the proportion of the white area and the color fading possibility evaluation index of the non-white area (monitoring area), and then classifies each specified clothing into clothing with the possibility of color fading and clothing without the possibility of color fading, and performs cleaning operations on different types of specified clothing respectively. This analysis method can more precisely evaluate the overall color fading risk of clothing. For different types of clothing, different cleaning strategies can be adopted to reduce the dyeing risk, which helps to protect the clothing and extend its service life.

[0047] It should be noted that when the present invention identifies clothing, it analyzes the possibility of color fading of specified clothing based on the collected images, and then classifies the specified clothing. This analysis method can intuitively and accurately understand the actual condition of the clothing and avoid the limitations of the material label information. The image-based analysis method is not limited by the type, material or label of the clothing, and can be applied to various different types of clothing, including clothing without labels or with incomplete label information, and can achieve fast and accurate clothing identification and classification, greatly improving the processing efficiency.

[0048] S5. Stain condition analysis: Locate the stain areas of each specified clothing based on the images of each specified clothing, and analyze the stain evaluation conditions of the stain areas of each specified clothing. The stain evaluation conditions specifically include stain color analysis and stain texture analysis.

[0049] In a preferred embodiment of the present invention, the stain color analysis needs to analyze the stain chromaticity values of the stain areas of each specified clothing. The specific method is as follows: Extract the images of the stain areas of each specified clothing, use image processing software to obtain the chromaticity values of the corresponding color areas of the stain areas of each specified clothing, construct the HSV color space of the corresponding color areas of the stain areas of each specified clothing, and then obtain the saturation and lightness of the corresponding color areas of the stain areas of each specified clothing. Calculate the average value of the saturation and lightness of the corresponding color areas of the stain areas of each specified clothing to obtain the monitoring chromaticity evaluation index of the corresponding color areas of the stain areas of each specified clothing.

[0050] Compare the monitoring chromaticity evaluation indexes of the corresponding color areas of the stain areas of each specified clothing, select the color area corresponding to the maximum monitoring chromaticity evaluation index as the characteristic color area of the stain area of the specified clothing, and obtain the chromaticity value of the characteristic color area as the stain chromaticity value of the stain area of the specified clothing.

[0051] It should be noted that the reason for selecting the color area corresponding to the maximum monitoring chromaticity evaluation index as the characteristic color area of the stain area of the specified clothing is as follows: 1. Saturation reflects the purity or vividness of the color. In stain analysis, the color area with a higher saturation may represent the core part of the stain or the original stain that has not been diluted or mixed. 2. Lightness determines the brightness or darkness of the color. In stain analysis, the change in the lightness value may reflect the contrast between the stain and the surrounding environment, as well as the brightness or darkness of the stain.

[0052] It should be explained that in the HSV color space, a set of data usually consists of three parameters: hue, saturation, and value. Exemplarily, if there is a set of data in the HSV color space as (H: 30°, S: 80%, V: 60%), this set of data represents a color with a hue of 30° (close to red but with a slight orange component), a saturation of 80% (the color is relatively vivid), and a value of 60% (medium brightness).

[0053] In a preferred embodiment of the present invention, the stain texture analysis needs to analyze the stain diffusivity index of each stain area of each specified clothing, and the specific method is as follows: Extract the images of each stain area of each specified clothing and identify the number of color areas corresponding to each stain area of each specified clothing.

[0054] If the number of color areas corresponding to a certain stain area of a certain specified clothing is 1, it is determined that the stain diffusivity index of this stain area of this specified clothing is 0.

[0055] It should be noted that if the number of color areas corresponding to a certain stain area of a certain specified clothing is 1, it indicates that there is no diffusion phenomenon in this stain area, and the color of this stain area is relatively uniform.

[0056] If the number of color areas corresponding to a certain stain area of a certain specified clothing is greater than 1, use image processing software to obtain the area of the innermost color area and the area of the outermost color area corresponding to each stain area of each specified clothing, and calculate the ratio of the area of the outermost color area to the area of the innermost color area corresponding to each stain area of each specified clothing to obtain the stain diffusivity index of each stain area.

[0057] It should be noted that the method for obtaining the area of the innermost color region and the area of the outermost color region: Obtain the stain regions of each specified piece of clothing, and then divide them according to the color regions to obtain the corresponding color regions of the stain regions of each specified piece of clothing. Obtain the center points of the corresponding stain regions, and then obtain the shortest distances from each color region to the center point, which are recorded as the monitoring distances. Compare the monitoring distances corresponding to each color region, select the color region corresponding to the minimum monitoring distance as the innermost color region, and select the color region corresponding to the maximum monitoring distance as the outermost color region. Then, use image processing software to obtain the area of the innermost color region and the area of the outermost color region corresponding to the stain regions of each specified piece of clothing.

[0058] S6. Stain type identification: Based on the stain evaluation of the stain regions of each specified piece of clothing, identify the stain types of the stain regions of each specified piece of clothing, and then select the corresponding cleaning methods and cleaning agents for different types of stains, so as to determine the cleaning methods and cleaning agents for different types of specified clothing.

[0059] In a preferred embodiment of the present invention, the method for identifying the stain types of the stain regions of each specified piece of clothing is as follows: Please refer to Figure 3 As shown, extract the stain chromaticity values and stain diffusivity indices of the stain regions of each specified piece of clothing, and then compare them with the pre-stored reference chromaticity value ranges and reference stain diffusivity index ranges corresponding to each stain type to identify the stain types of the stain regions of each specified piece of clothing.

[0060] In a preferred embodiment of the present invention, for selecting the corresponding cleaning methods and cleaning agents for different types of stains, it is necessary to identify the characteristic stain types of each specified piece of clothing: Extract the stain types of the stain regions of each specified piece of clothing, and then classify them to obtain the stain regions corresponding to each stain type of each specified piece of clothing. Use image processing software to obtain the area of each stain region corresponding to each stain type of each specified piece of clothing, and then perform cumulative calculation to obtain the stain area corresponding to each stain type of each specified piece of clothing.

[0061] Compare the stain areas corresponding to each stain type of each specified piece of clothing, and select the stain type corresponding to the largest stain area as the characteristic stain type of the specified piece of clothing.

[0062] It should be noted that different stain types require different cleaning agents and cleaning methods.

[0063] It should be further supplemented that the specific methods for determining the cleaning methods and cleaning agents for different types of designated clothing are as follows: extract the characteristic stain types of the designated clothing corresponding to the clothing with the possibility of color fading, and then count the quantities of the designated clothing corresponding to each characteristic stain type of the clothing with the possibility of color fading, and make a comparison. The stain type with the largest quantity of the designated clothing of the clothing with the possibility of color fading is used as the characteristic stain type of the clothing with the possibility of color fading.

[0064] Extract the characteristic stain types of the designated clothing corresponding to the clothing without the possibility of color fading, and then count the quantities of the designated clothing corresponding to each characteristic stain type of the clothing without the possibility of color fading, and make a comparison. The stain type with the largest quantity of the designated clothing of the clothing without the possibility of color fading is used as the characteristic stain type of the clothing without the possibility of color fading.

[0065] Select appropriate cleaning methods and cleaning agents based on the characteristic stain types of the clothing with the possibility of color fading and the characteristic stain types of the clothing without the possibility of color fading respectively.

[0066] It should be noted that when the present invention performs the cleaning of designated clothing, the stain type identification is carried out based on the analysis results of the stain color and stain texture of each stain area of each designated clothing. This analysis method can accurately identify the stain type based on the stain color and texture analysis. The washing and care robot can select the most suitable detergent and cleaning parameters for each type of stain, thereby improving the cleaning effect. The accurate stain type identification enables the washing and care robot to more effectively remove stains during a single cleaning process.

[0067] The above content is only an example and explanation of the concept of the present invention. Those skilled in the art of the present technology make various modifications or supplements or use similar methods to replace the specific embodiments described, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A method for clothes recognition and analysis by a washing robot based on image processing technology, characterized in that: include: S1. Capturing images of designated clothing: using a high-definition camera to capture images of designated clothing, wherein the images specifically include front and back images; S2. White area ratio analysis: analyzing the white area ratio of each designated clothing based on the image of each designated clothing; S3, analyzing the possibility of color fading of designated clothes: determining the monitoring area of ​​each designated clothes based on the image of each designated clothes, analyzing the color consistency and shape irregularity evaluation of each adjacent non-white area corresponding to the monitoring area of ​​each designated clothes, and then analyzing the possibility of color fading of each designated clothes; The analysis of the color consistency situation requires constructing a color consistency index for each designated clothing monitoring area corresponding to each adjacent non-white area; The analysis of the shape irregularity evaluation situation requires constructing a shape irregularity evaluation index corresponding to each adjacent non-white area of ​​each designated clothing monitoring area; The analysis of the fading possibility of each designated clothing item requires the construction of a fading possibility evaluation index for the monitoring area of ​​each designated clothing item, and the specific method is as follows: Extract the color consistency index and shape irregularity evaluation index of the monitoring area of ​​each designated clothing and each adjacent non-white area, and then sum them up according to the weight to obtain the color fading possibility evaluation index of the monitoring area of ​​each designated clothing; S4, classifying designated clothes: classifying each designated clothes based on the proportion of white area and the possibility of fading of each designated clothes to obtain the type of each designated clothes, wherein the type of each designated clothes specifically includes clothes with the possibility of fading and clothes without the possibility of fading, and then washing them respectively; S5, stain situation analysis: based on the image of each designated clothing, each stain area of ​​each designated clothing is located, and the stain evaluation situation of each stain area of ​​each designated clothing is analyzed, and the stain evaluation situation specifically includes stain color analysis and stain texture analysis; Stain color analysis requires analyzing the stain chromaticity of each stain area of ​​each specified garment; The stain texture analysis requires analyzing the stain diffusion index of each stain area of ​​each designated clothing; S6. Stain type identification: Identify the stain type of each stain area of ​​each designated clothing based on the stain evaluation of each stain area of ​​each designated clothing, and then select the corresponding cleaning method and detergent for different types of stains, and then determine the cleaning method and detergent for different types of designated clothing.

2. The method for identifying and analyzing clothing by a washing robot based on image processing technology as claimed in claim 1, characterized in that: The specific analysis method for analyzing the proportion of the white area of ​​each designated clothing is as follows: Extracting the front image and the back image of each designated clothing, and then using image processing software to obtain the white area of ​​the front image, the white area of ​​the back image, the clothing area of ​​the front image, and the clothing area of ​​the back image corresponding to each designated clothing; The white area ratio of the front image of each designated clothing item is calculated by the ratio of the white area area of ​​the front image to the clothing area of ​​the front image, and the white area ratio of the back image of each designated clothing item is calculated by the ratio of the white area area of ​​the back image of each designated clothing item to the clothing area of ​​the back image, to obtain the white area ratio of the back image of each designated clothing item; The white area ratio of the front image and the white area ratio of the back image of each designated clothing are averaged to obtain the monitored white area ratio of each designated clothing.

3. The method for clothes recognition and analysis by a washing robot based on image processing technology as claimed in claim 2, characterized in that: The specific method of constructing the color consistency index of each designated clothing monitoring area corresponding to each adjacent non-white area is as follows: Extract the front and back images of each designated clothing, and then use image processing software to obtain the area of ​​each non-white area of ​​each designated clothing, compare the areas of each non-white area of ​​each designated clothing, and select the non-white area corresponding to the largest area as the monitoring area; Using image processing software to obtain the chromaticity value of each designated clothing monitoring area and the chromaticity value of each corresponding adjacent non-white area; The chromaticity value of each designated clothing monitoring area and the chromaticity value of each corresponding adjacent non-white area are compared, and the color consistency index of each designated clothing monitoring area and each adjacent non-white area is analyzed and obtained.

4. The method for identifying and analyzing clothing by a washing robot based on image processing technology as claimed in claim 3, characterized in that: The specific method of constructing the shape irregularity evaluation index of each designated clothing monitoring area corresponding to each adjacent non-white area is as follows: The image processing software is used to obtain the area and perimeter of each adjacent non-white area corresponding to each designated clothing monitoring area, which are recorded as , ,in Indicates the number of the specified clothing. , Indicates the number of specified clothes. Indicates the number of the adjacent non-white area, , Indicates the number of adjacent non-white areas; Using the formula The shape irregularity evaluation index of each adjacent non-white area corresponding to each designated clothing monitoring area is obtained by analysis. ,in Indicates the preset reference area. Indicates a preset reference circumference.

5. The method for clothes recognition and analysis by a washing robot based on image processing technology as claimed in claim 1, characterized in that: The specific method of classifying each designated clothing is as follows: Extracting the monitored white area ratio and the monitored area fading possibility evaluation index of each designated clothing, and then comparing them with the preset monitored white area ratio threshold and the monitored area fading possibility evaluation index threshold respectively; If the proportion of the monitored white area of ​​a specified clothing item is less than the threshold of the monitored white area proportion and the monitoring area fading possibility assessment index is greater than the threshold of the monitoring area fading possibility assessment index, the specified clothing item is judged to be clothing with the possibility of fading; otherwise, the specified clothing item is judged to be clothing with no possibility of fading.

6. The method for clothes recognition and analysis by a washing robot based on image processing technology as claimed in claim 1, characterized in that: The stain color analysis needs to analyze the stain chromaticity of each stain area of ​​each designated clothing in the following specific manner: Extract the image of each stain area of ​​each designated clothing, use image processing software to obtain the chromaticity value of each color area corresponding to each stain area of ​​each designated clothing, construct the HSV color space of each color area corresponding to each stain area of ​​each designated clothing, and then obtain the saturation and brightness of each color area corresponding to each stain area of ​​each designated clothing, and calculate the mean of the saturation and brightness of each color area corresponding to each stain area of ​​each designated clothing to obtain the monitoring chromaticity evaluation index of each color area corresponding to each stain area of ​​each designated clothing; The monitored chromaticity evaluation indexes of each color area corresponding to each stain area of ​​each designated clothing are compared, and the color area corresponding to the maximum monitored chromaticity evaluation index is selected as the characteristic color area of ​​the stain area of ​​the designated clothing, and the chromaticity value of the characteristic color area is obtained as the stain chromaticity value of the stain area of ​​the designated clothing.

7. The method for clothes recognition and analysis by a washing robot based on image processing technology as claimed in claim 6, characterized in that: The specific method for analyzing the stain spread index of each stain area of ​​each designated garment is as follows: Extracting images of each stained area of ​​each designated clothing, and identifying the number of color areas corresponding to each stained area of ​​each designated clothing; If the number of color regions corresponding to a stain region of a specified clothing is 1, the stain diffusion index of the stain region of the specified clothing is judged to be 0; If the number of color areas corresponding to a stain area on a specified clothing is greater than 1, the image processing software is used to obtain the innermost color area area and the outermost color area area corresponding to each stain area of ​​each specified clothing, and the stain diffusion index of each stain area is calculated by calculating the ratio of the outermost color area area and the innermost color area area corresponding to each stain area of ​​each specified clothing.

8. The method for clothes recognition and analysis by a washing robot based on image processing technology as claimed in claim 7, characterized in that: The method of identifying the stain type of each stain area of ​​each designated clothing is as follows: The stain chromaticity value and stain diffusion index of each stain area of ​​each designated clothing are extracted, and then compared with the reference chromaticity value range and reference stain diffusion index range corresponding to each stain type stored in advance, so as to identify the stain type of each stain area of ​​each designated clothing.

9. The method for clothes recognition and analysis by a washing robot based on image processing technology as claimed in claim 8, characterized in that: The selection of the corresponding cleaning method and cleaning agent for different types of stains requires identifying the characteristic stain type of each specified clothing: Extract the stain type of each stain area of ​​each designated clothing, and then classify it to obtain the stain area corresponding to each stain type of each designated clothing, use image processing software to obtain the area of ​​each stain area corresponding to each stain type of each designated clothing, and then perform cumulative calculation to obtain the stain area corresponding to each stain type of each designated clothing; The stain areas corresponding to the stain types of the designated clothes are compared, and the stain type corresponding to the largest stain area is selected as the characteristic stain type of the designated clothes.

Images