Deep learning-based lobster online positioning visual detection method and system

Abstract

The invention provides a deep learning-based lobster online positioning visual detection method and system, and the method comprises the following steps: S1, obtaining a lobster image through the shooting of a visual recognition device, and transmitting the lobster image to an image processing platform to obtain a training data set; s2, identifying shrimp postures for the images in the training data set, and determining classification categories; s3, classifying the postures of the lobsters, finding out waist lines of decomposition of the heads and tails of the lobsters, and determining reference points of the waist lines; s4, the original images of the lobster postures under each classification form a sub-training data set, and the probability value of the area where each lobster waist line reference point is located under the classification is obtained; s5, bisecting the shrimp head, taking an angle bisector, and determining the angle direction of the shrimp head; s6, lobster positioning coordinates needing to be output are determined; and S7, sending to a control terminal based on the lobster positioning coordinates, and driving a finger clamp to grasp the lobster head and the lobster tail. The head and tail of the lobster can be automatically distinguished and identified, positioning is provided for subsequent head and tail separation operation, and the production efficiency is improved.

Description

Technical field [0001] The present invention relates to the field of lobster processing technology, in particular to a lobster online positioning visual detection method and system based on deep learning. Background [0002] At present, most of the operation of crayfish head removal in the market is done manually, not only the speed and efficiency of head removal are low, but also the health situation is not good, which is easy to cause secondary pollution of crayfish, which affects the quality of crayfish to a certain extent. [0003] Crayfish tail processing enterprises now urgently need an automated head removal equipment, with intelligent head removal equipment instead of labor, reduce the pollution of crayfish body in the head removal operation, and improve the quality and safety of crayfish tail, while increasing operational efficiency and reducing the production cost of crayfish tail. [0004] For example, the patent "a live skin crayfish head removal device" (201410475127...

AI Technical Summary

Problems solved by technology

This sorting link increases the process function of the production line, increases the cost, and reduces the overall efficiency

Images