Cashmere and wool identification method based on memristor neural network

• Main Authors: LU LIQUN, LI YUEFENG, TONG YI, SHI MINGMIN, ZHOU LYUYANG
• Format: Patent
• Language: Chinese; English
• Published: 25.10.2019
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

A cashmere and wool recognition method based on a memristor neural network comprises the following steps that S1, an original picture is input, the original picture is preprocessed through an input signal preprocessing circuit, and color information of the original picture is coded according to a memristor array specified signal source format; s2, the memristor neuron convolution network performsconvolution operation on the coded information according to the trained neural network to obtain an expected value; and S3, the expected value comparison and judgment circuit draws an expected value obtained by the memristor neuron convolution network into a phase sum degree curve, and compares the phase sum degree curve with an existing data curve in a database to obtain an identification result.The method can be used for detecting the authenticity of conventional wool and cashmere and analyzing the components of cashmere and wool products. The problems of long consumed time and large accuracy fluctuation of artifici