Image analysis method for evaluating pet hair state

Abstract

The invention provides an image analysis method for evaluating a pet hair state. The image analysis method comprises the following steps: acquiring pet hair images before and after using a hair care product by using an image acquisition device; performing image analysis on the pet hair image to obtain evaluation index data; and determining the pet hair state according to the evaluation index data. The method is a non-contact method, the original welfare of the animal is not damaged, the image acquisition equipment is mute in the whole process, the stress reaction of the pet is not caused, meanwhile, the method is suitable for acquisition in different scenes, and the test process is rapid, convenient and efficient.

Description

Technical Field

[0001] This application belongs to the technical field of pet daily chemical care products. Specifically, it relates to an image analysis method for evaluating the hair condition of pets. Background Art

[0002] Pets, also known as "companion animals", refer to various animals kept by families for the purposes of viewing, companionship, relieving people's mental stress or emotional sustenance. According to the data of the "2024 Development Report of the Chinese Pet Industry", with the continuous development of the pet market, the number of pets in China has gradually entered a relatively stable growth stage. It is estimated that the number of cats and dogs in 2024 will reach 187 million, with a year-on-year growth of about 1%. Among the types of pets raised in China, cats and dogs account for 74%, continuing to dominate the consumer market.

[0003] Nowadays, with the continuous penetration of the "parenting-style" pet-keeping concept, the needs of pet owners are tending to be more segmented and diversified. The data of the 2024 Pet Health Trend Insights Research by Tmall shows that hair health is one of the categories that pet owners pay more attention to, and pet owners have a high willingness to buy products to improve and maintain the hair health of their pets. According to the "2024 Pet Industry Insights Report of Xiaohongshu", pet hair beauty functional foods and hair care products have become the focus of attention of pet owners because they can improve the health and ornamental value of pet hair.

[0004] Hair care products refer to a category of products specifically designed for pets to clean, maintain, beautify or improve the hair health of pets. These products are usually developed according to the species of pets, hair types and specific needs to meet the special requirements of pet hair care, and have functions such as cleaning, maintenance, beautification, and health improvement, including shampoos, conditioners, dry shampoo sprays, emollient oils, etc.

[0005] Pet hair beauty functional foods refer to products such as foods, snacks, and nutritional supplements specifically designed for pets to improve and enhance the quality, color and growth of pet hair. These foods usually contain specific nutritional components such as proteins, fatty acids, vitamins and minerals.

[0006] The "2023 ByteDance Pet Industry White Paper" shows that in the field of beauty and grooming products, related content such as "knot removal" and "softness" has received much attention; data from Magic Mirror Insights (January - September 2024) shows that the functionality of dog shampoos has attracted much attention from consumers. Among them, hair care products have become the mainstay of the market with their characteristics of significantly improving the health and luster of pet hair, occupying a market share of 55.01%. It can be seen that pet owners are more concerned about the luster, combability, and softness of their pets' hair. These indicators can not only intuitively reflect the health of pet hair from different dimensions, but also help clinical researchers and pet product researchers judge the hair condition of pets, guide the development of products that better meet market trends and address consumers' pain points, and are also of great significance for the claim of hair care efficacy of products.

[0007] In recent years, some studies have evaluated the hair condition of pets by measuring pet hair-related indicators such as the peak force at hair breakage, cystine content, hair color difference value, and hair scale structure, and combining subjective sensory scores. Although these measurement methods can accurately and objectively obtain pet hair condition data, most of them require collecting ex vivo pet hair. The hair shaving process not only harms animal welfare but may also cause stress reactions in pets, and noisy hair shaving equipment may exacerbate the stress reaction; moreover, there are problems such as limited test sites, high costs, long cycles, and low efficiency. Summary of the Invention

[0008] The technical problem solved by this application is: how to more quickly, conveniently, and efficiently evaluate the hair condition of pets while avoiding harming animal welfare and causing stress reactions in pets.

[0009] This application provides an image analysis method for evaluating the hair condition of pets, and the image analysis method includes:

[0010] Using an image acquisition device to collect pet hair images before and after using hair care products;

[0011] Performing image analysis on the pet hair images to obtain evaluation index data;

[0012] Determining the hair condition of pets according to the evaluation index data.

[0013] Optionally, the method of using an image acquisition device to collect pet hair images before and after using hair care products includes:

[0014] Before using hair care products, using an image acquisition device to collect pet hair images of the first stage at a preset part;

[0015] After using hair care products, using an image acquisition device to collect pet hair images of the second stage at the preset part.

[0016] Optionally, the image analysis method further includes:

[0017] Repeatedly collect pet hair images at regular intervals to obtain several groups of pet hair images in the first stage and pet hair images in the second stage.

[0018] Optionally, the evaluation index data includes at least one of hair texture data, hair texture ratio data, hair glossiness data, and hair wrinkle data.

[0019] Optionally, the method for performing image analysis on the pet hair image to obtain evaluation index data includes:

[0020] Measure the roughness of the pet hair image to obtain the average roughness as the hair texture data.

[0021] Optionally, the method for performing image analysis on the pet hair image to obtain evaluation index data includes:

[0022] Perform spatial filtering on the pet hair image and adjust the image background and Gaussian radius to make the hair gaps prominent;

[0023] Measure and calculate the texture ratio area of the pet hair image to obtain the hair texture ratio data.

[0024] Optionally, the method for performing image analysis on the pet hair image to obtain evaluation index data includes:

[0025] Calculate the average intensity value in the color histogram corresponding to the pet hair image to obtain the hair glossiness data.

[0026] Optionally, the method for performing image analysis on the pet hair image to obtain evaluation index data includes:

[0027] Convert the pet hair image into a pseudo-color image and determine the hair wrinkle data according to the pseudo-color image.

[0028] Optionally, the method for determining the pet hair state according to the evaluation index data includes:

[0029] Analyze and statistically process the evaluation index data according to a preset algorithm to determine the significant difference degree of the pet hair change before and after using the hair care product;

[0030] Determine the pet hair state after using the hair care product according to the significant difference degree.

[0031] Optionally, the preset algorithm is a t-test method or a rank sum test method.

[0032] An image analysis method for evaluating the state of pet hair provided by the present application has the following technical effects:

[0033] This method is a non-contact method that does not harm animal welfare. The image acquisition device is completely silent throughout the process and will not cause stress reactions in pets. At the same time, it is suitable for collection in different scenarios, and the testing process is fast, convenient, and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a main step flow chart of an image analysis method for evaluating the state of pet hair according to one or more embodiments;

[0035] Figure 2 It is a graph showing the change results of hair texture at time points T0 and T1 according to one or more embodiments;

[0036] Figure 3 It is a graph showing the change results of hair wrinkles at time points T0 and T1 according to one or more embodiments. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0038] Before describing the various embodiments of the present application in detail, the technical concept of the present application will be briefly described first: Currently, when measuring pet hair, pet hair is usually collected ex vivo. The collection process is likely to damage animal welfare and cause stress reactions in pets, and there are also problems such as limited testing sites, high costs, long cycles, and low efficiency. For this reason, the image analysis method for evaluating the state of pet hair provided by the present application uses an image acquisition device to collect pet hair images before and after using hair care products, and combines the use of corresponding algorithms to analyze and count the pet hair images to determine the state of pet hair. This method is a non-contact method that does not damage animal welfare. The image acquisition device is completely silent throughout the process and will not cause stress reactions in pets. At the same time, it is suitable for collection in different scenarios, and the testing process is fast, convenient, and efficient. The specific principle of the image analysis method for evaluating the state of pet hair of the present application will be described below with more embodiments.

[0039] Specifically, as Figure 1 shown, the image analysis method for evaluating the state of pet hair in this embodiment includes the following steps:

[0040] Step S10: Use an image acquisition device to collect pet hair images before and after using hair care products.

[0041] Step S20: Perform image analysis on the pet hair image to obtain evaluation index data.

[0042] Step S30: Determine the pet hair condition according to the evaluation index data.

[0043] In one or more embodiments, the method of using an image acquisition device to acquire pet hair images before and after using a hair care product in step S10 includes: before using the hair care product, using the image acquisition device to acquire the first-stage pet hair image of a preset part; after using the hair care product, using the image acquisition device to acquire the second-stage pet hair image of the preset part. Among them, the hair care product includes hair care products and beauty functional foods.

[0044] Exemplarily, the preset parts include the middle of the back (the part where the highest points of the two scapulas of the animal intersect with the dorsal midline), the left and right front legs (the hair on the humerus part of the animal), and the waist (the part where the hip bone intersects with the dorsal midline). Image acquisition device. Exemplarily, the image acquisition device includes a high-definition digital camera and an Antera 3D skin imaging analyzer.

[0045] When acquiring the first-stage pet hair image before using the hair care product, it is necessary to use a straight comb to comb the pet's hair, and under the conditions of a fixed light source and a fixed position, use a high-definition digital camera and Antera 3D to acquire a part image with a size of not less than 4×5 cm. 2 When acquiring the second-stage pet hair image after using the hair care product, first conduct a trial of the hair care product: according to the product usage requirements, conduct a product trial on the pet. After the product trial, it is necessary to ensure that the hair is in a dry state and use a straight comb to comb the hair. Then, under the conditions of a fixed light source and a fixed position, use a high-definition digital camera and Antera 3D to acquire a part image with a size of not less than 4×5 cm. 2 Exemplarily, repeat the acquisition of pet hair images at regular intervals to obtain several groups of first-stage pet hair images and second-stage pet hair images. During the interval, it is necessary to feed the pet food, feed the pet at a fixed time and fixed weight (set according to the product characteristics), and the feeding cycle and image acquisition cycle are set according to the actual situation.

[0046] Exemplarily, the evaluation index data includes at least one of hair texture data, hair texture proportion data, hair glossiness data, and hair wrinkle data. Among them, the average roughness is obtained by measuring the roughness of the pet hair image as the hair texture data. The pet hair image is processed with a spatial filter, and the image background and Gaussian radius are adjusted to highlight the analysis object; the proportion area of the texture of the analysis object is measured and calculated to obtain the hair texture proportion data. The average intensity value in the color histogram corresponding to the pet hair image is calculated to obtain the hair glossiness data. The pet hair image is converted into a pseudo-color image, and the hair wrinkle data is determined according to the pseudo-color image.

[0047] After obtaining the evaluation index data, the evaluation index data is analyzed and statistically processed according to a preset algorithm to determine the significant difference degree of the pet hair change before and after using the hair care product, and the pet hair state after using the hair care product is determined according to the significant difference degree. Among them, the preset algorithm is the t-test method or the rank sum test method.

[0048] The field of pet hair testing is an emerging field in response to the rapid development of the pet industry in recent years. Different from economic animals, companion animals mainly play the role of companionship. In the process of humans taking care of pets, emotional satisfaction is obtained by observing and touching the pets, and more attention is paid to the appearance / welfare of the pets. In this embodiment, the hair indexes of the pet's fur before and after using / feeding the product are detected by an image acquisition device to evaluate the effect of the product on the pet's hair.

[0049] The innovation of this method is to apply the human local skin detection instrument and equipment across fields to the hair of pets, overcoming the technical prejudice against the use of the instrument. The method does not harm animal welfare, the image acquisition device is completely silent throughout the process and will not cause stress reactions in pets; the economic cost is low, the test efficiency is high, and it is more objective than the single subjective sensory scoring method (vision, touch), and there are few such non-contact methods in the current pet testing field. The change of the pet state or the evaluation of the use effect of the washing and care products / beautiful hair foods can be intuitively presented through image analysis.

[0050] The evaluation indexes selected by this method are more in line with the market trend and can better hit the pain points of consumers compared with the original group standard. It can guide R & D personnel to design product development with the goal of pet health. The indexes selected by the method can correspond to the specific efficacy of the product, which is more conducive to the development of foods / washing and care products that promote the health of pet hair than the original group standard.

[0051] This method has a wide range of applications and can be extended to other fields: The image acquisition device is highly portable and is widely applicable to scenarios such as families, pet grooming shops, product R & D centers, etc. It effectively avoids the stress reactions that may be caused by domestic pets changing the environment / going to the laboratory environment, helps to establish pet personalized files / pet health management, track the state of hair at different times, and verify the claimed effects of grooming products / beautiful hair foods, etc.

[0052] The following uses multiple embodiments to separately illustrate the specific process of this image analysis method.

[0053] Embodiment 1: Obtain hair texture data to evaluate the softness of pet hair.

[0054] Test steps: Select a part of 4×6 cm on the pet's body, and use Antera 3D to take the pet hair image at the first stage (measurement time point T0), ensuring that the angle and parameters are consistent. Use a cleaning product to clean the pet as a whole, and after drying / blowing dry, use the imaging device to take the pet hair image at the second stage (measurement time point T1). 2 Analysis of the image collected by Antera 3D: Use the corresponding supporting software to analyze the hair texture index. The specific principle is determined by measuring the average roughness Ra using the roughness plane correction option.

[0055] Data analysis: Use special statistical software such as SAS or SPSS, EXCEL, etc. for statistics, and use t or rank sum test (if the data is normally distributed, use the t - test method for data analysis and statistics; if the data is non - normally distributed, use the rank sum test method for statistics).

[0056] Test results: Before and after using the cleaning product, the change results of hair texture at time points T0 and T1 are significantly different (p < 0.05), indicating that the product can improve the softness of hair. See Table 1 for details.

[0057] 。 Figure 2 。

[0058] Table 1 Analysis table of hair texture change results before and after using the product

[0059]

[0060]

[0061] Embodiment 2: Obtain the proportion data of hair texture to evaluate the softness of pet hair.

[0062] Test steps: Select a part of 4×6 cm on the pet's body 2For the selected area, take the first-stage pet hair image (measurement time point T0) with a high-definition digital camera, ensuring consistent angles and parameters. Use a cleaning product to clean the entire pet, and after drying / blowing dry, use the imaging device to take the second-stage pet hair image (measurement time point T1).

[0063] Analysis of the images collected by the high-definition digital camera: Use Image-pro Plus software to analyze the hair texture ratio index. This analysis uses spatial filters, sets the background, Gaussian radius, etc., to make the hair gaps prominent, and measures and calculates the area of the texture ratio.

[0064] Data analysis: Use dedicated statistical software such as SAS, SPSS, or EXCEL for statistics, and use t-tests or rank sum tests (if the data is normally distributed, use the t-test method for data analysis and statistics; if the data is non-normally distributed, use the rank sum test method for statistics).

[0065] Test results: There are significant differences in the changes in the texture ratio at the T0 and T1 time points before and after using the cleaning product (p < 0.05), indicating that the product can improve hair softness. See Table 2 for details.

[0066] Table 2 Analysis table of the changes in the texture ratio before and after using the product

[0067]

[0068]

[0069] Examples 1 and 2 prove that using the non-contact test method of image analysis to measure two indicators, namely hair texture and texture ratio, can be applied to the actual operation of evaluating pet hair softness.

[0070] Example 3: Obtain hair gloss data to evaluate pet hair gloss.

[0071] Test steps: Select an area of 4×6 cm on the pet's body 2 For the selected area, take the first-stage pet hair image (measurement time point T0) with a high-definition digital camera, ensuring consistent angles and parameters. Use a cleaning product to clean the entire pet, and after drying / blowing dry, use the imaging device to take the second-stage pet hair image (measurement time point T1).

[0072] Analysis of the images collected by the high-definition digital camera: Use Image-pro Plus software to analyze the glossiness index. This analysis observes the average value of Intensity in the color histogram, observes its changes, and evaluates the change in hair brightness.

[0073] Data analysis: Use dedicated statistical software such as SAS, SPSS, or EXCEL for statistics. Adopt t-test or rank sum test (if the data is normally distributed, use the t-test method for data analysis and statistics; if the data is non-normally distributed, use the rank sum test method for statistics).

[0074] Test results: Before and after using the cleaning product, there are significant differences in the change results of glossiness at time points T0 and T1 (p < 0.05), indicating that the product can enhance the glossiness of pet hair. See Table 3 for details.

[0075] Table 3 Analysis table of glossiness change results before and after using the product

[0076]

[0077] Example 3 proves that using the non-contact test method of image analysis to measure the glossiness index of pet hair can be applied to the actual operation of pet hair glossiness evaluation.

[0078] Example 4: Obtain hair wrinkle data to evaluate the combability of pet hair.

[0079] Test steps: Select a 4×6 cm area on the pet's body, and use Antera 3D to take the pet hair image at the first stage (measurement time point T0), ensuring that the angle and parameters are consistent. Use the cleaning product to clean the pet as a whole, and after drying / blowing dry, use the imaging device to take the pet hair image at the first stage again (measurement time point T1). 2 Image analysis of Antera 3D acquisition: Use the corresponding supporting software to analyze the hair combability index. The specific principle is to convert the color image into a pseudo-color image, where dark red represents the deepest wrinkled feature, and white corresponds to the top of the surface.

[0080] Data analysis: Use dedicated statistical software such as SAS, SPSS, or EXCEL for statistics. Adopt t-test or rank sum test (if the data is normally distributed, use the t-test method for data analysis and statistics; if the data is non-normally distributed, use the rank sum test method for statistics).

[0081] Test results: Before and after using the cleaning product, there are significant differences in the change results of hair combability at time points T0 and T1 (p < 0.05), indicating that the product has the effect of making the hair easier to comb. See

[0082] Table 4 for details. Figure 3 Table 4 Analysis table of combability change results before and after using the product

[0083] Table 4 Analysis table of combability change results before and after using the product

[0084]

[0085] Example 4 proves that the use of the non-contact test method of image analysis to measure the hair wrinkle index can be applied to the actual operation of pet hair combability evaluation.

[0086] Examples 1, 2, 3, and 4 above prove that using the non-contact test method to measure the four indicators of hair texture / texture ratio, gloss, and hair wrinkle can be faster, more convenient,

[0087] and more efficient in evaluating the state of pet hair, and is widely applicable to scenarios such as families, pet grooming shops, and product R & D centers.

[0088] The specific embodiments of the present application have been described in detail above. Although some embodiments have been shown and described, those skilled in the art should understand that these embodiments can be modified and improved without departing from the principle and spirit of the present application defined by the claims and their equivalents, and these modifications and improvements should also be within the protection scope of the present application.

Claims

1. An image analysis method for evaluating the condition of pet hair, characterized in that: The image analysis method comprises: Using an image acquisition device to capture images of pet hair before and after using a hair care product; Performing image analysis on the pet hair image to obtain evaluation index data; The pet's hair condition is determined according to the evaluation index data.

2. The image analysis method for evaluating the condition of pet hair according to claim 1, characterized in that: The method for collecting pet hair images before and after using the hair care product using an image acquisition device comprises: Before using the hair care product, using an image acquisition device to acquire a first-stage pet hair image of a preset part; After using the hair care product, the second stage pet hair image of the preset part is collected by using the image acquisition device.

3. The image analysis method for evaluating the condition of pet hair according to claim 2, characterized in that: The image analysis method further comprises: The pet hair images are repeatedly collected at predetermined intervals to obtain a plurality of groups of first-stage pet hair images and second-stage pet hair images.

4. The image analysis method for evaluating the condition of pet hair according to claim 1, characterized in that: The evaluation index data includes at least one of hair texture data, hair texture ratio data, hair glossiness data and hair wrinkle data.

5. The image analysis method for evaluating the condition of pet hair according to claim 4, characterized in that: The method for performing image analysis on the pet hair image to obtain evaluation index data comprises: The roughness of the pet hair image is measured to obtain an average roughness as hair texture data.

6. The image analysis method for evaluating the condition of pet hair according to claim 4, characterized in that: The method for performing image analysis on the pet hair image to obtain evaluation index data comprises: Performing spatial filter processing on the pet hair image, and adjusting the image background and Gaussian radius to highlight the hair gaps; The texture ratio of the hair image is measured and calculated to obtain hair texture ratio data.

7. The image analysis method for evaluating the condition of pet hair according to claim 4, characterized in that: The method for performing image analysis on the pet hair image to obtain evaluation index data comprises: The average intensity value in the color histogram corresponding to the pet hair image is calculated to obtain hair glossiness data.

8. The image analysis method for evaluating the condition of pet hair according to claim 4, characterized in that: The method for performing image analysis on the pet hair image to obtain evaluation index data comprises: The pet hair image is converted into a pseudo color image, and hair wrinkle data is determined based on the pseudo color image.

9. The image analysis method for evaluating the condition of pet hair according to claim 1, characterized in that: The method for determining the pet hair status according to the evaluation index data includes: Analyzing and collecting the evaluation index data according to a preset algorithm to determine the significant difference in the pet's hair changes before and after using the hair care product; The condition of the pet's hair after using the hair care product is determined based on the degree of significant difference.

10. The image analysis method for evaluating the condition of pet hair according to claim 9, characterized in that: The preset algorithm is a t-test method or a rank-sum test method.

Images