Novel macrobrachium phenotype data measurement and classification method

Abstract

The invention relates to a new macrobrachium phenotype data measurement and classification method, which comprises the following steps of: putting macrobrachium into a transparent box filled with water, sending the transparent box into macrobrachium picture acquisition equipment, and starting a camera to obtain a top view and a side view of the macrobrachium. The method has the beneficial effects that the phenotype data of the macrobrachium can be integrally, accurately and quickly obtained by adopting a computer vision method of deep learning and stereoscopic vision, the body length, the second-step foot length, the color and other information of the macrobrachium can be obtained with relatively high precision, and high-throughput measurement and classification of the phenotype data of the macrobrachium are realized through data analysis; according to the method, the complexity and low efficiency of manual measurement and classification in the original method are overcome; the method is based on high-integration-level integrated computer software and model, and operation is simple and efficient; the invention provides a method for measuring the second-step foot length of macrobrachium based on a space coordinate relation for the first time; a conjoint analysis method of macrobrachium phenotype data and step color is provided for the first time, and macrobrachium population analysis is realized.

Description

technical field [0001] The invention belongs to the field of computer vision based on deep learning, and in particular relates to a new method for measuring and classifying giant macrobrach phenotype data. Background technique [0002] Macrobrachium has the advantages of wide appetite, fast growth, delicious meat, and high economic value. Since my country introduced it in 1976, it has carried out large-scale promotion and breeding in coastal provinces. At present, China has become the country with the largest number of Macrobrachium rosenbergii farming in the world, and the farming output accounts for more than 60% of the world's total output. During the breeding period, the population of Macrobrachium prawns in the culture ponds was regularly morphologically measured and statistically analyzed. According to the ratio of the length of the second leg of the shrimp to the total length of the shrimp body and the color of the second leg, the Macrobrachium prawn was divided into ...

AI Technical Summary

Problems solved by technology

[0008] 1. Traditional manual measurement methods are time-consuming and labor-intensive, and error-prone

[0009] 2. The above-mentioned method with detection and segmentation as the main technology does not raise the accuracy to a higher level. In the detection task of Macrobrachium prawns, the segmentation model of shrimp body has complicated shrimp whiskers and shrimp feet, etc., and the effect is poor.

[0010] 3. The model based on key point detection currently has no mature integrated technology applied to the data measurement and analysis of Macrobrachium

[0011] 4. Macrobrachium population classification is mostly based on manual experience

Images