Intelligent Detection Methods of Fresh Corn Ear Maturity Based on Texture

摘要

Partial image of fresh corn ear was collected by quality analysis detection system based on computer vision technology, and the texture information related to the ear maturity was extracted to complete maturity intelligent detection. Gray gradient co-occurrence matrix was set up by ear image after gray transformation. Then 15 texture features was extracted and reduced dimensions by the calculation of the correlation coefficient and principal component analysis. The first and the second principal component were used as inputs of probabilistic neural network, developed for intelligent detection of fresh corn ear maturity, with detection accuracy 88.89%.