A method for evaluating animal fur cutting effect

Abstract

A method for evaluating the cutting effect of animal fur, wherein the fur contour image is obtained by subjecting the animal fur to image denoising and image contour extraction; the sharp parts of the fur are removed, and the parts to be cut are defined as the ideal invalid area, and the parts to be retained are defined as the ideal effective area. The area of ​​the region is expressed by the sum of the number of white pixels in the region, and the ideal invalid area is composed of sharp areas such as limbs that need to be cut, and the ideal effective area is the area retained after cutting, and the area retained by the final cutting, and the remaining waste edge area is calculated at the same time; the final score is judged by comparing the effective area with the invalid area; the cutting path diagram obtained by the actual cutting is compared with the cutting diagram of the ideal state by using the scoring of the effective area and the invalid area to judge the quality of the cutting effect. The present invention can clearly judge whether the cutting effect is good or bad, thereby optimizing the automatic cutting technology to meet the subsequent industrial requirements of automatic cutting of animal fur.

Description

Technical Field

[0001] The invention relates to the technical field of animal fur cutting, and in particular to an animal fur cutting effect evaluation method. Background Art

[0002] Fur cutting is the primary processing step in the leather industry. Currently, the cutting of large raw hides is labor-intensive and enterprises lack new young labor, which leads to difficulties in recruiting workers. Automatic fur cutting can not only improve the production and processing efficiency of enterprises and optimize their operating costs, but also relieve the pressure of recruiting and employing workers.

[0003] Currently, most fur cutting is done manually, and automated fur cutting has not yet become popular. In addition, there are currently no corresponding cutting indicators or standards for automated fur cutting. Therefore, a method is needed to provide a standard for automated fur cutting, so as to quantitatively evaluate the cutting effect. Summary of the invention

[0004] In order to overcome the shortcomings of the above-mentioned prior art, the purpose of the present invention is to provide a method for evaluating the cutting effect of animal fur, which can evaluate the cutting results of animal fur images of different shapes and sizes and give corresponding scores, and can provide corresponding cutting effect reference indicators for the automated cutting industry. Through the scoring evaluation of this method, it is possible to clearly judge whether the cutting effect is good or bad, thereby optimizing the automatic cutting technology to meet the subsequent industrial requirements of automated cutting of animal fur.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A method for evaluating the cutting effect of animal fur comprises the following steps:

[0007] Step (1): performing image denoising and image contour extraction on the animal fur to obtain a fur contour image;

[0008] Step (2): According to the method of cutting fur in the factory, the sharp parts of the fur are removed, and the parts that need to be cut are defined as the ideal invalid area Sm, and the parts that need to be retained are defined as the ideal effective area Sn. The area of ​​the area is expressed by the sum of the number of white pixels in the area, where S = Sm + Sn. The ideal invalid area is composed of sharp areas such as limbs that need to be cut. At this time, Sm = S1 + S2 + S3 + S4. The ideal effective area is the area retained after cutting. The area retained by the final cutting is calculated as Snn', and the area overlapping with the ideal effective area Sn is the actual effective area Snn. At the same time, the remaining waste edge area Smm' is calculated, Smm' = S-Snn', and the area overlapping with the ideal invalid area Sm is the actual invalid area Smm;

[0009] Step (3): Use the formula As an evaluation index of fur cutting effect, this formula can be used to judge the final score by comparing the effective area with the ineffective area. The higher the Q value, the better the cutting effect. The best cutting effect Q value is 100%.

[0010] Step (4): Use this formula to compare the cutting path diagram obtained by actual cutting with the cutting diagram of the ideal state by using the scores of effective area and invalid area to obtain the final score, so as to judge the quality of the cutting effect.

[0011] In the step (2), the actual effective area Snn and the actual ineffective area Smm represent the overlapping area of ​​the image after actual cutting and the image after ideal cutting.

[0012] Beneficial effects of the present invention:

[0013] The method of the present invention can evaluate the effect of cutting animal fur, provide an evaluation standard for the research and development of automated cutting technology, evaluate the results of cutting animal fur images of different shapes and sizes, and give corresponding scores, and provide corresponding cutting effect reference indicators for the automated cutting industry. Through the score evaluation of the method, it is possible to clearly judge whether the cutting effect is good or bad, thereby optimizing the automated cutting technology to meet the subsequent industrial requirements for automated cutting of animal fur. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention automates the step of cutting the fur.

[0015] Figure 2 It is the research object of the present invention.

[0016] Figure 3 It is a schematic diagram of the effective area and the ineffective area of ​​the present invention.

[0017] Figure 4 Figure 2. Actual effective / ineffective area diagram of the present invention.

[0018] Figure 5 Figure 1 is a step diagram for evaluating the final effect of the present invention. DETAILED DESCRIPTION

[0019] The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0020] Prepare the materials:

[0021] The general processing steps for automatic fur cutting are roughly divided into: image denoising, image contour extraction, and image cutting line path planning. Figure 1 The figure shows the automatic fur cutting process flow chart. The final result extracted in the main control center (such as a computer) is the path cutting diagram on the contour image, such as Figure 1 Therefore, in view of this feature, this paper takes the contour image as the research object and formulates the standard, such as Figure 2 As shown, typical fur contour images were selected as the research objects for standard setting.

[0022] Standard formulation basis:

[0023] The key to automatic fur cutting is to cut off the waste fur in the edge areas such as the head, hoof, and tail according to the characteristics of the fur edge, so as to complete the edge trimming of the raw hide, reduce the waste of chemical raw materials in the subsequent tanning process, and facilitate the secondary processing of the leather. Therefore, the parts that need to be cut have distinct characteristics, so it can be judged based on the cutting situation on the industrial site that most cutting paths are approximately straight lines to cut off the protruding sharp parts such as the limbs.

[0024] Cutting judgment criteria:

[0025] Based on a large number of on-site pictures of fur cutting in factories, it can be seen that most fur is mainly removed from sharp parts, so the part that needs to be cut can be defined as the ideal invalid area Sm, and the part that needs to be retained can be defined as the ideal effective area Sn. The area of ​​the area is expressed by the sum of the number of white pixels in the area, and the overall contour area of ​​the fur is calculated using Python software to obtain the total area S, as shown in Figure 3 (a), where S = Sm + Sn, and the ideal ineffective area consists of sharp areas such as the limbs that need to be cut, such as Figure 3 As shown in (b), at this time, Sm = S1 + S2 + S3 + S4, and the ideal effective area is the area retained after cutting, such as Figure 3 (c) as shown.

[0026] The area retained by the final cutting is calculated as Snn', and the area overlapping with the ideal effective area Sn is the actual effective area Snn. At the same time, the remaining scrap area Smm' is calculated, Smm' = S-Snn', and the area overlapping with the ideal invalid area Sm is the actual invalid area Smm. Figure 4 As shown, the actual effective area Snn and the actual ineffective area Smm represent the overlapping area of ​​the image after actual cutting and the image after ideal cutting.

[0027] According to the actual situation, an empirical formula (1) is formulated as an evaluation index for judging the fur cutting effect. The formula (1) is shown in the figure below:

[0028]

[0029] This formula can be used to judge the final score by comparing the effective area with the ineffective area. The higher the Q value, the better the cutting effect. The best cutting effect Q value is 100%. In this state, the effective area retained by the actual cutting is equal to the ideal effective area, and the ineffective area removed by the actual cutting is equal to the ideal ineffective area removed, that is, Snn = Sn, Smm = Sm. The final evaluation of the cutting effect can be carried out according to the score.

[0030] Score evaluation:

[0031] According to the image cutting line experimental result diagram obtained after actual cutting, Figure 5 The effect evaluation shown in the figure is to calculate the corresponding areas of the actual cutting and the ideal cutting according to the Q index. The image size used is 5 million (2448×2048) pixels. The corresponding areas calculated are shown in Table 1, where all areas are replaced by the number of white pixels in units. According to formula (1), Q=90.81%. The comprehensive use of the score comparison of the effective area and the invalid area can effectively avoid the situation of mis-segmentation.

[0032] Table 1 Image evaluation analysis table

[0033]

[0034]

Claims

1. A method for evaluating the effect of animal fur cutting, characterized in that: The steps include: Step (1): performing image denoising and image contour extraction on the animal fur to obtain a fur contour image; Step (2): According to the method of cutting fur in the factory, the sharp parts of the fur are removed, and the parts that need to be cut are defined as the ideal invalid area Sm, and the parts that need to be retained are defined as the ideal effective area Sn. The area of ​​the area is expressed by the sum of the number of white pixels in the area, where S = Sm + Sn. The ideal invalid area is composed of the sharp areas of the limbs that need to be cut. At this time, Sm = S1 + S2 + S3 + S4. The ideal effective area is the area retained after cutting. The area retained by the final cutting is calculated as Snn', and the area overlapping with the ideal effective area Sn is the actual effective area Snn. At the same time, the remaining waste edge area Smm' is calculated, Smm' = S-Snn', and the area overlapping with the ideal invalid area Sm is the actual invalid area Smm; Step (3): Use the formula As an evaluation index of fur cutting effect, this formula can be used to judge the final score by comparing the effective area with the ineffective area. The higher the Q value, the better the cutting effect. The best cutting effect Q value is 100%. Step (4): Use this formula to compare the cutting path diagram obtained by actual cutting with the cutting diagram of the ideal state by using the scores of effective area and invalid area to obtain the final score, so as to judge the quality of the cutting effect.

Images