Multi-Feature Recognition of Healthy Vegetable Seedlings Based on Machine Vision Technology

The quality of potted seedlings has an important influence on the yield vegetables during seedling raising and transplanting. inconsistency after transplanting is main factor causing decline in vegetable yield. To eliminate or reduce this influence, health test before particularly to ensure crop In study, image recognition technology based machine vision proposed. It a multi-feature method for non-destructive detection healthy seedlings. color pot enhanced by industrial control computer system self-written algorithm (hereinafter referred as SIXA algorithm). segmentation denoising are realized ultra-green threshold 3D Block Matched filtering (BM3D) algorithm. Information about leaf area features was collected. criteria confirmed analyzed. Among them, feature thresholds study were set R≥60.7; G≥119.4; B≥1.9, F≥0.15. This limitation identifying single information establish identification seedlings, aiming improve accuracy experimental verification shows that overall rate platform high 96.67%, which meets expectations.