ABSTRACT
The quality of peanut kernels is referred to every aspect of the profit of supply and marketing. A back propagation (BP) neural network model of quality grade testing and identification is built based on 52 appearance features such as the shape, texture, and color using the technology of computer image processing. A comprehensive test is carried out in 1,400 grains to find out unsound kernel, mildewing, impurity, hetero-variety, and other aspects with the aim of achieving results with the accuracy rate of 95.6%. According to the national standards, a method is designed for testing peanut kernels’ grade based on their specification and quality, in which 100 peanut grains are tested with results of the comprehensive test reaching the accuracy rate of 92%. The methods that are discussed this chapter to test the quality and distinguish the grade of peanuts based on their appearance and specifications can produce results with high accuracy rate, which can have a positive impact on the peanut industry’s productivity and development.
