yolov5的iou
时间: 2023-09-07 20:14:20 浏览: 150
yolov5
YOLOv5中的IOU(Intersection over Union)是一种衡量检测模型性能的指标。它是通过计算预测框与真实标注框之间的重叠面积与它们的并集面积之比来衡量的。当IOU值超过设定的IOU阈值时,认为该预测框与真实标注框匹配成功,否则认为匹配失败。 [2]
在YOLOv5中,有两个与IOU相关的参数。
1. mAP@0.5:当IOU为0.5时的mAP(mean Average Precision),它是每个类别的AP值的平均数。而AP表示平均精确率(Precision)在不同召回率(Recall)下的变化曲线下的面积,用于评估多类别标签预测的性能。mAP@0.5越高,模型的性能越好。<span class="em">1</span><span class="em">2</span><span class="em">3</span>
#### 引用[.reference_title]
- *1* *3* [YOLOv5的置信度阀值与iou阀值及P R详解](https://blog.csdn.net/excelNo1/article/details/123871655)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v93^chatsearchT3_2"}}] [.reference_item style="max-width: 50%"]
- *2* [Yolov5——评估指标](https://blog.csdn.net/REstrat/article/details/126873627)[target="_blank" data-report-click={"spm":"1018.2226.3001.9630","extra":{"utm_source":"vip_chatgpt_common_search_pc_result","utm_medium":"distribute.pc_search_result.none-task-cask-2~all~insert_cask~default-1-null.142^v93^chatsearchT3_2"}}] [.reference_item style="max-width: 50%"]
[ .reference_list ]
阅读全文