Areca nut detection method based on neural network algorithm

Abstract

The invention provides an areca nut detection method based on a neural network algorithm, and the method comprises a training stage comprising obtaining a training image data set, marking the trainingimage data set to construct a training sample, inputting the training sample into a convolutional neural network, obtaining a connection weight and an offset value of the convolutional neural network, and obtaining a convolutional neural network training model; a detection stage, comprising: acquiring an actual areca nut image as input into the convolutional neural network training model to obtain an areca nut detection result. According to the invention, the VGG convolutional neural network is used as a classifier for areca nut defect detection. Since the network structure is shared by the convolutional neural network weight, the complexity of a network structure is reduced, the number of weights is reduced, so that the classification speed is high. In addition, samples with suspected defects are used for training, the convolutional neural network automatically learns the complex class condition density of the samples, the problem caused by artificial assumption of a class conditiondensity function is avoided, and the detection precision is improved.

Description

technical field [0001] The invention belongs to the field of image target recognition, and in particular relates to a betel nut visual detection method based on a deep convolutional neural network algorithm. Background technique [0002] At present, the betel nut products in the domestic market are mainly concentrated in Hunan Province. The traditional betel nut sorting process is labor-intensive, low in production efficiency, low in mechanization level, and high in labor costs. Along with the research and optimization to the betel nut processing procedure, various betel nut seed selection machines have also occurred at present, disclose a kind of betel nut screening equipment as Chinese utility model patent CN207086366U. When the betel nut screening equipment is implemented, the betel nuts to be screened are transported through the betel nut conveying track, the light-emitting device emits detection light and irradiates the betel nuts to be screened on the betel nut conveyi...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problem that the traditional visual algorithm classification method is unable to classify non-standard parts (betel nuts), propose a kind of betel nut detection method based on neural network algorithm, realize the sorting of betel nut raw seeds, and can effectively remove waste seeds, Improve sorting efficiency and stability and reliability, improve the level of automation and intelligence, and reduce manufacturing and labor costs

Images