基于机器视觉和机器学习技术的浙贝母外观品质等级区分

董成烨,李东方,冯槐区,龙思放,奚特,周芩安,王俊

Classification of Fritillaria thunbergii appearance quality based on machine vision and machine learning technology

Chengye DONG,Dongfang LI,Huaiqu FENG,Sifang LONG,Te XI,Qin’an ZHOU,Jun WANG

图4 YOLO-X的预测效果A～F.单目标图像检测结果；G～J.多目标图像检测结果。图片下方红色叉代表模型检测时发生误检或漏检。标签中数字为模型检测浙贝母类别时的置信度，图5同。

Fig. 4 Prediction results by YOLO-XA-F. Single object image detection results; G-J. Multi-object image detection results. The red crosses at the bottom of the image represent false detections or missed detections during the model detection. The number in the label is the confidence of the model when detecting the class of F. thunbergii, and the same as in Fig. 5.