A novel method for pest detection is proposed based on the theory of multi-fractal spectrum to extract pests on plant leaves.Both local and global information of the image regularity were obtained by multi-fractal ana...A novel method for pest detection is proposed based on the theory of multi-fractal spectrum to extract pests on plant leaves.Both local and global information of the image regularity were obtained by multi-fractal analysis.By applying fractal dimension,the spots on leaf images can be extracted without loosing or introducing any information.The different parts of images are re-analysis by morphology operations to determine the candidate regions of pests.The performance of multi-fractal analysis of whitefly detection is investigated through greenhouse experiments.The experimental results show that the proposed method is robust to noise from light and is not sensitive to the complex environment.展开更多
A novel approach is proposed to automatically detect pomographic images with skin-like color background on the Intemet using the locations of human faces and bodies. It has two separate skin-color detection steps: th...A novel approach is proposed to automatically detect pomographic images with skin-like color background on the Intemet using the locations of human faces and bodies. It has two separate skin-color detection steps: the first one is to quickly detect the potential human skin-color regions; and the second one is to use an off-the-shelf face detector to locate a human face and then apply hypothesis testing based on series of assumptions which take into account the face-height ratio, body orientation and modem photograph composition common sense, etc. After that, a template matching method is used to further discriminate normal images or pornographic ones. Experimental results show that the proposed method has high precision and real time speed.展开更多
本研究提出了一套基于深度学习的图像目标检测以及机械臂抓取技术的自动化核酸采样系统,采用“机械臂+摄像头”的方式对入境物品及其包装进行图像目标检测并进行拭子采样。本研究的目标检测框架以基于YOLOv5框架为蓝本,使用深度可分离...本研究提出了一套基于深度学习的图像目标检测以及机械臂抓取技术的自动化核酸采样系统,采用“机械臂+摄像头”的方式对入境物品及其包装进行图像目标检测并进行拭子采样。本研究的目标检测框架以基于YOLOv5框架为蓝本,使用深度可分离卷积以及逐点卷积融合的方式作为新的主干网络,设计了更加轻量级的幻影卷积进行Neck结构中的特征提取,以此来改进检测速度。最后,通过比例-积分-微分(Proportion Integral Differential,PID)来控制机械臂进行对应物体的抓取或涂抹操作,以此保证机械臂在运行过程中的稳定性。在自建数据集上本研究修改后的模型检测指标达到了99.18%,在CPU上使用ONNX框架部署的速度达到了25 FPS,能够满足实时检测的需求。展开更多
基金supported by the MKE(The Ministry of Knowledge Economy),Korea,under the ITRC(Infor mationTechnology Research Center)support programsupervised by the NIPA(National IT Industry Promotion Agency)(NIPA-2010-C1090-1021-0010)
文摘A novel method for pest detection is proposed based on the theory of multi-fractal spectrum to extract pests on plant leaves.Both local and global information of the image regularity were obtained by multi-fractal analysis.By applying fractal dimension,the spots on leaf images can be extracted without loosing or introducing any information.The different parts of images are re-analysis by morphology operations to determine the candidate regions of pests.The performance of multi-fractal analysis of whitefly detection is investigated through greenhouse experiments.The experimental results show that the proposed method is robust to noise from light and is not sensitive to the complex environment.
文摘A novel approach is proposed to automatically detect pomographic images with skin-like color background on the Intemet using the locations of human faces and bodies. It has two separate skin-color detection steps: the first one is to quickly detect the potential human skin-color regions; and the second one is to use an off-the-shelf face detector to locate a human face and then apply hypothesis testing based on series of assumptions which take into account the face-height ratio, body orientation and modem photograph composition common sense, etc. After that, a template matching method is used to further discriminate normal images or pornographic ones. Experimental results show that the proposed method has high precision and real time speed.
文摘本研究提出了一套基于深度学习的图像目标检测以及机械臂抓取技术的自动化核酸采样系统,采用“机械臂+摄像头”的方式对入境物品及其包装进行图像目标检测并进行拭子采样。本研究的目标检测框架以基于YOLOv5框架为蓝本,使用深度可分离卷积以及逐点卷积融合的方式作为新的主干网络,设计了更加轻量级的幻影卷积进行Neck结构中的特征提取,以此来改进检测速度。最后,通过比例-积分-微分(Proportion Integral Differential,PID)来控制机械臂进行对应物体的抓取或涂抹操作,以此保证机械臂在运行过程中的稳定性。在自建数据集上本研究修改后的模型检测指标达到了99.18%,在CPU上使用ONNX框架部署的速度达到了25 FPS,能够满足实时检测的需求。