KNN-based bighead carp classification method

Abstract

The invention belongs to the technical field of image recognition and classification, and particularly relates to a KNN-based bighead carp classification method, which comprises the following steps ofA1, collecting image data of a plurality of bighead carps, and storing the image data in a background database; A2, classifying the bighead carp image data in the background database into reservoir growth bighead carps and pond culture bighead carps, and dividing the reservoir growth bighead carps and the pond culture bighead carps into a training set and a test set; A3, converting the pluralityof image data in the background database into vectors, and standardizing the image data; A4, taking the training set pictures as input of a convolutional neural network, and performing feature extraction; and A5, constructing a KNN classifier to classify the bighead carps. According to the KNN-based bighead carp classification method, different morphological characteristics of bighead carps in twogrowth environments can be extracted, and then the bighead carps are identified and classified according to the difference of the extracted morphological characteristics.

Description

technical field [0001] The invention belongs to the technical field of image recognition and classification, and in particular relates to a KNN-based bighead carp classification method. Background technique [0002] The reservoir water purification fishery industry has good ecological and environmental protection benefits, social public benefits and industrial economic benefits. Compared with fish farmed in ponds, fish in reservoirs are free-range fish in reservoirs with better quality, higher nutrition and better taste, so the price is much higher than fish farmed in ponds. However, in the fish market, there is a phenomenon that reservoir fish and farmed fish are mixed, and it is pond farmed fish that consumers tend to buy at high prices. In order to protect the rights and interests of consumers and solve consumers' doubts about the quality of reservoir fish and aquatic products, a big data traceability system for reservoir fish is needed. When consumers purchase reservoir...

AI Technical Summary

Problems solved by technology

Among the above-mentioned existing technologies, no researchers have used image recognition and classification technology to identify and classify the differences in morphological characteristics of the same species of fish due to their growth in different environments, and they have applied it to fish products. In the traceability system

Images