Rice Quality Analyzer

1.application areas: processing plant, storage warehouse, quality inspection center;

2. Detection scope: broken rice grains, insect erosion grains, disease spot grains, mold grains, grass seeds, chalky grains, brown rice grains, impurities, etc;

3. Key technologies:

1) Automatic binarization: The use of deep neural network to segment the foreground background of the image, compared with the traditional two-value method, can be applied to a variety of lighting conditions, and.rice edge segmentation is smoother, which has the advantages of fast speed and high robustness.

2) Adhesion Rice Segmentation Algorithm: The method based on the connected domain cannot divide the adhesion.rice, the use of deep neural network to the adhesion of rice instance-level segmentation, can achieve.1000fpsspeed, can be processed in real time on the adhesion of rice.

3) Rice attribute recognition algorithm: Using lightweight neural networks and integrating semi-supervised learning methods, the model can be iteratively optimized with only a small amount of data labeled by the user, with the advantages of high precision, high speed and convenient deployment.