Abstract: A target detection method based on fusion of vision, lidar and millimeter wave radar comprises: obtaining original data detected by a camera, a millimeter wave radar, and a lidar, and synchronizing the millimeter wave radar, the lidar, and the camera in time and space; performing a calculation on the original data detected by the millimeter wave radar according to a radar protocol; generating a region of interest by using a position, a speed, and a radar reflection area obtained from the calculation; extracting feature maps of a point cloud bird's-eye view and the original data detected by the camera; projecting the region of interest onto the feature maps of the point cloud bird's-eye view and the original data detected by the camera; fusing the feature maps of the point cloud bird's-eye view and the original data detected by the camera, and processing a fused image through a fully connected layer.

Abstract: Provided is a traffic sign recognition method based on a lightweight neural network, which including: a lightweight neural network model is constructed for training and pruning to obtain a lightweight neural network model; the lightweight neural network model comprises a convolution feature extraction part and a classifier part; the convolution feature extraction part includes one layer of conventional 3×3 convolution and 16 layers of separable asymmetric convolution. The classifier part includes three layers of separable full connection modules.

Abstract: Provided is a traffic sign recognition method based on a lightweight neural network, which including: a lightweight neural network model is constructed for training and pruning to obtain a lightweight neural network model; the lightweight neural network model comprises a convolution feature extraction part and a classifier part; the convolution feature extraction part includes one layer of conventional 3×3 convolution and 16 layers of separable asymmetric convolution. The classifier part includes three layers of separable full connection modules.

Abstract: Disclosed are a defense method and a model of deep learning model aiming at adversarial attacks in the technical field of image recognition, which makes full use of the internal relationship between the adversarial samples and the initial samples, and transforms the adversarial samples into common samples by constructing a filter layer in front of the input layer of the deep learning model; the parameters of the filter layer are trained by using the adversarial attack samples, so as to improve the ability of the model to resist adversarial attack; then the trained filter layer is combined with the learning model after the adversarial training, and a deep learning model with strong robustness and high classification accuracy is obtained, which ensures that the recognition ability of the initial sample is not reduced while resisting the adversarial attacks.

Abstract: Disclosed are a point cloud polar coordinate coding method and a device, including dividing a circular scanning area scanned by a lidar at an equal angle with an angle ?? to obtain a plurality of identical polar coordinate areas; dividing each of the polar coordinate areas with equal length along a radial direction with a length ?r to obtain a plurality of polar coordinate grids and generating a plurality of polar coordinate cylinders corresponding to each of the polar coordinate grids in a three-dimensional space; generating polar coordinate cylinder voxels; extracting structural features from the all point cloud data in each of the polar coordinate cylinder voxels; obtaining a two-dimensional point cloud pseudo-image; boundary supplementing to the two-dimensional point cloud pseudo-image; and performing feature extraction on the two-dimensional point cloud pseudo-image by using convolutional neural networks, and outputting a final feature map.

Abstract: A target detection method based on fusion of vision, lidar and millimeter wave radar comprises: obtaining original data detected by a camera, a millimeter wave radar, and a lidar, and synchronizing the millimeter wave radar, the lidar, and the camera in time and space; performing a calculation on the original data detected by the millimeter wave radar according to a radar protocol; generating a region of interest by using a position, a speed, and a radar reflection area obtained from the calculation; extracting feature maps of a point cloud bird's-eye view and the original data detected by the camera; projecting the region of interest onto the feature maps of the point cloud bird's-eye view and the original data detected by the camera; fusing the feature maps of the point cloud bird's-eye view and the original data detected by the camera, and processing a fused image through a fully connected layer.