wheat quality analyzer

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

2. Detection scope: sprouts, xenogeneic sprouts, grass seeds, insect-etched grains, erythema grains, damaged grains, black embryo grains, lesion 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.wheat edge segmentation is smoother, which has the advantages of fast speed and high robustness.

2) Adhesion wheat segmentation algorithm: The method based on the connected domain cannot divide the adhesion.wheat, the use of deep neural network to the adhesion of wheat instance-level segmentation, can achieve.1000fpsspeed, can real-time processing of adhesion of wheat.

3) wheat 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.