Nanjing Nongda Cotton Grade Visual Recognition Technology Breakthrough

please

For a long time, cotton picking and purchasing were mainly organoleptic test and instrumental testing was supplementary, and the quality of cotton picking needs to be improved.

Since 2005, the Wang Ling team has begun to study the recognition and classification of cotton seed in the field based on image feature selection.

According to the introduction, the researchers first used the CCD digital camera to collect the front image of the three stage cotton seed cotton "Su Mian 12" at the early stage, the middle stage and the late stage, and the front image of 350 grade cotton seeds. Then, through the technical treatment, the seed cotton image was segmented from the complex background, and the 14 texture characteristics reflecting the seed cotton whiteness, yellow and impurity content and the 16 shape features reflecting the size and structure of the cotton petal were used to classify and classify the seed cotton, and the recognition rates reached 85.3% and 88.4%, respectively, for the cotton seed 1~7.

Associate professor Wang Ling said that the cotton picking robot equipped with visual identification technology can simulate artificial real-time picking cotton seed and grading, which can adapt to the variety of cotton varieties, reduce production costs and improve cotton grade conformity rate, so that the collection of high, medium and low grade cotton will become more reasonable.

The breakthrough of this technology will also change the cotton quality test in our country from post harvest detection to pre harvest detection. Through picking robot, we can know before that the quality of cotton in which field is good and the quality of cotton in the field is poor, so as to avoid duplication or meaningless labor and realize the refinement of cotton picking.

It is learnt that at present, China has not yet fully adapted to the national conditions of the cotton grading equipment. For this reason, with the cotton picking robot vision system in seed cotton maturity and grade characterization research in depth, will also be conducive to the national conditions of cotton.