LIDAR point cloud-based 3D object detection aims to sense the surrounding environment by anchoring objects with the Bounding Box(BBox).However,under the three-dimensional space of autonomous driving scenes,the previou...LIDAR point cloud-based 3D object detection aims to sense the surrounding environment by anchoring objects with the Bounding Box(BBox).However,under the three-dimensional space of autonomous driving scenes,the previous object detection methods,due to the pre-processing of the original LIDAR point cloud into voxels or pillars,lose the coordinate information of the original point cloud,slow detection speed,and gain inaccurate bounding box positioning.To address the issues above,this study proposes a new two-stage network structure to extract point cloud features directly by PointNet++,which effectively preserves the original point cloud coordinate information.To improve the detection accuracy,a shell-based modeling method is proposed.It roughly determines which spherical shell the coordinates belong to.Then,the results are refined to ground truth,thereby narrowing the localization range and improving the detection accuracy.To improve the recall of 3D object detection with bounding boxes,this paper designs a self-attention module for 3D object detection with a skip connection structure.Some of these features are highlighted by weighting them on the feature dimensions.After training,it makes the feature weights that are favorable for object detection get larger.Thus,the extracted features are more adapted to the object detection task.Extensive comparison experiments and ablation experiments conducted on the KITTI dataset verify the effectiveness of our proposed method in improving recall and precision.展开更多
A series of oxidants supported on coconut shell-based activated carbon(CAC) through microwave irradiation were prepared and characterized using scanning electron microscopy(SEM), N_2 adsorption/desorption analysis, an...A series of oxidants supported on coconut shell-based activated carbon(CAC) through microwave irradiation were prepared and characterized using scanning electron microscopy(SEM), N_2 adsorption/desorption analysis, and X-ray photoelectron spectroscopy(XPS). The SO_2 adsorption capacities and rates were evaluated by adsorption tests performed in a fixed bed reactor with a simulated flue gas, and the adsorption isotherm models were validated against the experimental results. The findings revealed that the SO_2 adsorption capacity decreased in the following order: MW-K_2Cr_2O_7-CAC > MWKMnO_4-CAC > MW-H_2O_2-CAC > MW-CAC. The SO_2 adsorption capacities and adsorption rates of the samples increased with an increasing oxidizability of the oxidants owing to the increment of mean pore size and oxygen-containing functional groups. In addition, a high initial SO_2 concentration and a low bed temperature could positively affect the SO2 adsorption. Finally, the Langmuir model validated that SO_2 was mainly adsorbed through chemical adsorption on the sample surfaces.展开更多
基金This work was supported,in part,by the National Nature Science Foundation of China under grant numbers 62272236in part,by the Natural Science Foundation of Jiangsu Province under grant numbers BK20201136,BK20191401in part,by the Priority Academic Program Development of Jiangsu Higher Education Institutions(PAPD)fund.
文摘LIDAR point cloud-based 3D object detection aims to sense the surrounding environment by anchoring objects with the Bounding Box(BBox).However,under the three-dimensional space of autonomous driving scenes,the previous object detection methods,due to the pre-processing of the original LIDAR point cloud into voxels or pillars,lose the coordinate information of the original point cloud,slow detection speed,and gain inaccurate bounding box positioning.To address the issues above,this study proposes a new two-stage network structure to extract point cloud features directly by PointNet++,which effectively preserves the original point cloud coordinate information.To improve the detection accuracy,a shell-based modeling method is proposed.It roughly determines which spherical shell the coordinates belong to.Then,the results are refined to ground truth,thereby narrowing the localization range and improving the detection accuracy.To improve the recall of 3D object detection with bounding boxes,this paper designs a self-attention module for 3D object detection with a skip connection structure.Some of these features are highlighted by weighting them on the feature dimensions.After training,it makes the feature weights that are favorable for object detection get larger.Thus,the extracted features are more adapted to the object detection task.Extensive comparison experiments and ablation experiments conducted on the KITTI dataset verify the effectiveness of our proposed method in improving recall and precision.
文摘A series of oxidants supported on coconut shell-based activated carbon(CAC) through microwave irradiation were prepared and characterized using scanning electron microscopy(SEM), N_2 adsorption/desorption analysis, and X-ray photoelectron spectroscopy(XPS). The SO_2 adsorption capacities and rates were evaluated by adsorption tests performed in a fixed bed reactor with a simulated flue gas, and the adsorption isotherm models were validated against the experimental results. The findings revealed that the SO_2 adsorption capacity decreased in the following order: MW-K_2Cr_2O_7-CAC > MWKMnO_4-CAC > MW-H_2O_2-CAC > MW-CAC. The SO_2 adsorption capacities and adsorption rates of the samples increased with an increasing oxidizability of the oxidants owing to the increment of mean pore size and oxygen-containing functional groups. In addition, a high initial SO_2 concentration and a low bed temperature could positively affect the SO2 adsorption. Finally, the Langmuir model validated that SO_2 was mainly adsorbed through chemical adsorption on the sample surfaces.
