为提高小麦病害检测精度,实现将模型方便快速部署到移动端,该研究提出了一种基于改进YOLOv8的轻量化小麦病害检测方法。首先,使用PP-LCNet模型替换YOLOv8网络结构的骨干网络,并在骨干网络层引入深度可分离卷积(depthwise separable conv...为提高小麦病害检测精度,实现将模型方便快速部署到移动端,该研究提出了一种基于改进YOLOv8的轻量化小麦病害检测方法。首先,使用PP-LCNet模型替换YOLOv8网络结构的骨干网络,并在骨干网络层引入深度可分离卷积(depthwise separable convolution, DepthSepConv)结构,减少模型参数量,提升模型检测性能;其次,在颈部网络部分添加全局注意力机制(global attention mechanism, GAM)模块,强化特征中语义信息和位置信息,提高模型特征融合能力;然后,引入轻量级通用上采样内容感知重组(content-aware reassembly of features,CARAFE)模块,提高模型对重要特征的提取能力;最后,使用Wise-IoU(weighted interpolation of sequential evidence for intersection over union)边界损失函数代替原损失函数,提升网络边界框回归性能和对小目标病害的检测效果。试验结果表明,对于大田环境下所采集的小麦病害数据集,改进后模型的参数量及模型大小相比原YOLOv8n基线模型分别降低了12.5%和11.3%,同时精确度(precision)及平均精度均值(mean average precision,m AP)相较于原模型分别提高了4.5和1.9个百分点,优于其他对比目标检测算法,可为小麦病害检测无人机等移动端检测装备的部署和应用提供参考。展开更多
Bioceramics have attracted extensive attention for bone defect repair due to their excellent bioactivity and degradability.However,challenges remain in matching the rate between bioceramic degradation and new bone for...Bioceramics have attracted extensive attention for bone defect repair due to their excellent bioactivity and degradability.However,challenges remain in matching the rate between bioceramic degradation and new bone formation,necessitating a deeper understanding of their degradation properties.In this study,density functional theory(DFT)calculations was employed to explore the structural and electronic characteristics of silicate bioceramics.These findings reveal a linear correlation between the maximum isosurface value of the valence band maximum(VBM_(Fmax))and the degradability of silicate bioceramics.This correlation was subsequently validated through degradation experiments.Furthermore,the investigation on phosphate bioceramics demonstrates the potential of this descriptor in predicting the degradability of a broader range of bioceramics.This discovery offers valuable insights into the degradation mechanism of bioceramics and holds promise for accelerating the design and development of bioceramics with controllable degradation.展开更多
文摘为提高小麦病害检测精度,实现将模型方便快速部署到移动端,该研究提出了一种基于改进YOLOv8的轻量化小麦病害检测方法。首先,使用PP-LCNet模型替换YOLOv8网络结构的骨干网络,并在骨干网络层引入深度可分离卷积(depthwise separable convolution, DepthSepConv)结构,减少模型参数量,提升模型检测性能;其次,在颈部网络部分添加全局注意力机制(global attention mechanism, GAM)模块,强化特征中语义信息和位置信息,提高模型特征融合能力;然后,引入轻量级通用上采样内容感知重组(content-aware reassembly of features,CARAFE)模块,提高模型对重要特征的提取能力;最后,使用Wise-IoU(weighted interpolation of sequential evidence for intersection over union)边界损失函数代替原损失函数,提升网络边界框回归性能和对小目标病害的检测效果。试验结果表明,对于大田环境下所采集的小麦病害数据集,改进后模型的参数量及模型大小相比原YOLOv8n基线模型分别降低了12.5%和11.3%,同时精确度(precision)及平均精度均值(mean average precision,m AP)相较于原模型分别提高了4.5和1.9个百分点,优于其他对比目标检测算法,可为小麦病害检测无人机等移动端检测装备的部署和应用提供参考。
基金National Key Research and Development Program of China (2023YFB3813000)National Natural Science Foundation of China (52272256)State Key Laboratory of Advanced Technology for Materials Synthesis and Processing (Wuhan University of Technology)(2022-KF-77)。
文摘Bioceramics have attracted extensive attention for bone defect repair due to their excellent bioactivity and degradability.However,challenges remain in matching the rate between bioceramic degradation and new bone formation,necessitating a deeper understanding of their degradation properties.In this study,density functional theory(DFT)calculations was employed to explore the structural and electronic characteristics of silicate bioceramics.These findings reveal a linear correlation between the maximum isosurface value of the valence band maximum(VBM_(Fmax))and the degradability of silicate bioceramics.This correlation was subsequently validated through degradation experiments.Furthermore,the investigation on phosphate bioceramics demonstrates the potential of this descriptor in predicting the degradability of a broader range of bioceramics.This discovery offers valuable insights into the degradation mechanism of bioceramics and holds promise for accelerating the design and development of bioceramics with controllable degradation.
