An automatic visual detection method for double-yellow duck eggs based on distance contours

Abstract

The invention discloses an automatic visual inspection method for two-yolk duck eggs based on a distance contour line, and relates to the field of agricultural product non-destructive testing technology. The device is as follows: a light source is arranged in the center of the bottom of a dark box, a to-be-detected duck egg is placed right above the light source, an industrial camera is placed right above the to-be-detected duck egg, and the industrial camera is connected with a computer through a data transmission line. The method comprises the following steps: (1) collecting duck egg images;(2) pretreating the duck egg images; (3) extracting egg yolk characteristics; (4) judging whether two yolks exist in a duck egg or not. Whether two yolks exist in each duck egg is independently analyzed, and the method is independent of large sample statistics analysis; erosion and dilation for grey-scale maps are adopted for extraction of yolk region characteristics, and the threshold value is automatically taken, so that the obtained yolk contour is more accurate; double-yolk eggs and single-yolk eggs are distinguished by using the contour line, and the method is more visual and accurate; the method is high in robustness and wide in application range.

Description

technical field [0001] The invention relates to the technical field of non-destructive detection of agricultural products, in particular to an automatic visual detection method for double-yellow duck eggs based on distance contour lines. Background technique [0002] On the one hand, due to the rich nutritional value of double-yolk eggs and people’s consumer psychology towards the good meaning of double-yolk eggs, it has become a hot-selling item in the consumer market, and its commercial value is far greater than that of single-yolk duck eggs; on the other hand, Double-yolk eggs cannot be hatched as breeding eggs, so double-yolk eggs should be removed when selecting breeding eggs. Therefore, the non-destructive identification of double yolk eggs is particularly important. [0003] At present, my country's small and medium-sized poultry egg production enterprises mostly use manual candling to distinguish double-yolk eggs from single-yolk eggs, which is labor-intensive, susc...

AI Technical Summary

Problems solved by technology

However, there may be some problems with this method: 1. The patent believes that the shape of the yolk of a double-yolk egg is a double circle. In actual processing, the obtained yolk image may be irregular; 2. Directly use Otsu to segment the yolk. If there is interference, Generally, the segmented area is relatively large; 3. Using neural network training requires a large number of samples, and the accuracy rate cannot be guaranteed to be 100%. Different varieties of duck eggs need multiple trainings

However, this method needs to analyze a large number of samples to ensure the correctness of the results.

Images