Fire detection has a great impact on people’s life safety.Fire Detection-DETR(FD-DETR)is a fire detection model based on RT-DETR for early fire identification in complex fire scenes.In this study,Adown sub-sampling m...Fire detection has a great impact on people’s life safety.Fire Detection-DETR(FD-DETR)is a fire detection model based on RT-DETR for early fire identification in complex fire scenes.In this study,Adown sub-sampling module was selected to improve the original convolution module,which improved the detection accuracy and reduced the number of parameter values.Using LSKA attention module on the backbone network further improved the detection accuracy.The experimental results showed that compared with the original RT-DETR model,the precision and mAP of FD-DETR flame detection are increased by 0.8%and 0.1%,respectively,which proves that the improved method proposed in this study effectively improves the feature extraction and feature fusion capabilities of the network.In the complex scene fire detection task,the performance of the improved RT-DETR algorithm is better than the original RT-DETR algorithm.展开更多
为了实现光照变化等复杂环境下果实的选择性采摘,该研究以黄瓜为研究对象,以RT-Detr为基线网络,提出了RT-Detr-EV模型。首先在主干网络中添加RepVGG模块,以加强网络特征提取能力,并减少推理时计算量;加入轻量化自注意力机制,减少计算量...为了实现光照变化等复杂环境下果实的选择性采摘,该研究以黄瓜为研究对象,以RT-Detr为基线网络,提出了RT-Detr-EV模型。首先在主干网络中添加RepVGG模块,以加强网络特征提取能力,并减少推理时计算量;加入轻量化自注意力机制,减少计算量,增加网络深度;最后使用MPDIoU(minimum point distance based intersection over union)替换原模型中的损失函数,加快模型的收敛,提高模型的检测准确率。研究表明,改进RT-Detr-EV的平均精度均值mAP50相较于原模型提升了3.2个百分点,检测速度相较原模型提升了17.4帧/s。与YOLOv7-X、YOLOv8-l相比,对非适宜采摘的黄瓜识别准确率分别提升4.6、6.5个百分点,检测速度分别提升了40.6、25帧/s,参数量分别减少了55.5%、27.3%。同时试验证明,模型对光照条件多种变化的采摘场景也具有一定的鲁棒性与泛化能力。该研究提出的RT-Detr-EV模型能够满足复杂生长环境黄瓜果实的实时检测需求,可为后续移动式选择性采摘的研究提供技术支持。展开更多
文摘Fire detection has a great impact on people’s life safety.Fire Detection-DETR(FD-DETR)is a fire detection model based on RT-DETR for early fire identification in complex fire scenes.In this study,Adown sub-sampling module was selected to improve the original convolution module,which improved the detection accuracy and reduced the number of parameter values.Using LSKA attention module on the backbone network further improved the detection accuracy.The experimental results showed that compared with the original RT-DETR model,the precision and mAP of FD-DETR flame detection are increased by 0.8%and 0.1%,respectively,which proves that the improved method proposed in this study effectively improves the feature extraction and feature fusion capabilities of the network.In the complex scene fire detection task,the performance of the improved RT-DETR algorithm is better than the original RT-DETR algorithm.
文摘为了实现光照变化等复杂环境下果实的选择性采摘,该研究以黄瓜为研究对象,以RT-Detr为基线网络,提出了RT-Detr-EV模型。首先在主干网络中添加RepVGG模块,以加强网络特征提取能力,并减少推理时计算量;加入轻量化自注意力机制,减少计算量,增加网络深度;最后使用MPDIoU(minimum point distance based intersection over union)替换原模型中的损失函数,加快模型的收敛,提高模型的检测准确率。研究表明,改进RT-Detr-EV的平均精度均值mAP50相较于原模型提升了3.2个百分点,检测速度相较原模型提升了17.4帧/s。与YOLOv7-X、YOLOv8-l相比,对非适宜采摘的黄瓜识别准确率分别提升4.6、6.5个百分点,检测速度分别提升了40.6、25帧/s,参数量分别减少了55.5%、27.3%。同时试验证明,模型对光照条件多种变化的采摘场景也具有一定的鲁棒性与泛化能力。该研究提出的RT-Detr-EV模型能够满足复杂生长环境黄瓜果实的实时检测需求,可为后续移动式选择性采摘的研究提供技术支持。
