Search

In autonomous driving, lidar has become the main vehicle sensor due to its advantages such as long-range measurement and high accuracy. However, the collected point cloud data is sparse and unevenly distributed, and it lacks characterization capabilities when facing objects with missing or similar shapes, so that the detection accuracy is low while detecting long-distance small targets with similar shapes and a small number of point clouds. In order to improve the detection accuracy of small targets represented by point clouds, this paper adopts a method that fuses point cloud and RGB image to construct a 3D object detection network architecture based on two-stage complementary fusion. In the first stage of the fusion, we use the FPS method to select some points from the raw point cloud data as key points. Then, we voxelize the raw point cloud and use the 3D sparse convolutional neural network to extract multi-scale points cloud features, which would fuse features of different scales with key points. In the second stage of fusion, a 2D object detector is used to obtain the 2D bounding box and category information of the target in the image and take the camera as the origin to extend along the direction of the 2D bounding box to form a frustum; then the point cloud and target category information within the frustum are fused into the key points. This paper uses key points as a bridge to effectively combine the semantic information of the image such as texture and color with the point cloud features. The experimental results show that the network proposed in this paper has excellent performance on small objects. [ABSTRACT FROM AUTHOR]

Irony recognition is an important research direction in text sentiment analysis, which contributes to discover ironic tone to judge text emotion correctly. The object of this paper is the application of deep learning to irony recognition, and the research aims to solve the problem that existing machine learning algorithms have difficulty in discriminating ironic tones. Aiming at the irony recognition of Chinese social comments, DC-BiGRU-CNN is proposed from the perspective of structural optimization rather than a text vectorization mechanism point of view, which is a dual-channel CNN combined with BiGRU, and incorporates attention mechanism and multigranularity convolutional neural network as the main framework. This paper first briefly discusses the difficulties of irony recognition and provides an overview of existing text sentiment analysis algorithms. This is followed by a detailed discussion of DC-BiGRU-CNN. Furtherly, it is compared with the main irony recognition methods on a social comment dataset containing Chinese ironic comments, and the experimental results show that DC-BiGRU-CNN can improve the accuracy of irony recognition. [ABSTRACT FROM AUTHOR]

In order to solve the declining influence of traditional cultural symbols, the research on traditional cultural symbols has become more meaningful. This article aims to study the application of traditional cultural symbols in art design under the background of artificial intelligence. In this paper, a fractal model with self-combined nonlinear function changes is constructed. By combining nonlinear transformations and multiparameter adjustments, various types of fractal models can be automatically rendered. The convolutional neural network algorithm is used to extract the characteristics of the style picture, and it is compared with the trained picture many times to avoid the problem of excessive tendency of the image with improper weight. The improved L-BFGS algorithm is also used to optimize the loss of the traditional L-BFGS, which improves the quality of the generated pictures and reduces the noise of the chessboard. The experimental results in this paper show that the improved L-BFGS algorithm has the least loss and the shortest time in the time used for more than 500 s. Compared with the traditional AdaGrad method, its loss is reduced by about 62%; compared with the traditional AdaDelta method, its loss is reduced by 46%. Its loss is reduced by about 8% compared with the newly optimized Adam method, which is a great improvement. [ABSTRACT FROM AUTHOR]

Convolution neural network has become a hot research topic in the field of computer vision because of its superior performance in image classification. Based on the above background, the purpose of this paper is to analyze sports sequence images based on convolutional neural network. In view of the low detection rate of single-frame and the complexity of multiframe detection algorithms, this paper proposes a new algorithm combining single-frame detection and multiframe detection, so as to improve the detection rate of small targets and reduce the detection time. Based on the traditional residual network, an improved, multiscale, residual network is proposed in this paper. The network structure enables the convolution layer to "observe" data from different scales and obtain more abundant input features. Moreover, the depth of the network is reduced, the gradient vanishing problem is effectively suppressed, and the training difficulty is reduced. Finally, the ensemble learning method of relative majority voting is used to reduce the classification error rate of the network to 3.99% on CIFAR-10, and the error rate is reduced by 3% compared with the original residual neural network. [ABSTRACT FROM AUTHOR]

With the rapid development of China's electrical industry, the safe operation of electrical facilities is very important for social stability and people's property safety. The failure detection method of conventional electrical equipment is hand detection, which has high experience of the detection person, lacks detection and error detection, and the detection efficiency is low. With the development of artificial intelligence technology, computer-assisted substation inspection is now possible, and substation inspection using an intelligent inspection robot equipped with an infrared device is one of the main substation inspection methods. In this paper, experiments are carried out using several neural network models. For example, if a faster region convolutional neural networks (RCNN) infrared detection model is employed, a good vg16 in the feature region of the extracted image takes into account the quality of the infrared image and the presence of multiple devices. Infrared images can be used to determine the basic features of various electronic devices. In order to detect targets in infrared images of electrical equipment, the fast RCNN target detection algorithm is used, and the overall recognition accuracy reaches 83.1%, and a good application effect is obtained. [ABSTRACT FROM AUTHOR]

There are many research studies in the field of breast cancer prediction, but it has been observed that the time taken for prediction needs to be reduced. The problem in the existing research is space consumption by graphical content. The proposed research is supposed to minimize the prediction time and space consumption. In this paper, research has focused on the study of existing breast cancer research and techniques and eliminating their limitation. It has been observed that when the number of datasets increases, every comparison makes a huge gap in size and comparison time. This research proposes a methodology for breast cancer prediction using an edge-based CNN (convolutional neural network) algorithm. The elimination of useless content from the graphical image before applying CNN has reduced the time consumption along with space consumption. The edge detection mechanism would retail only edges from the image sample in order to detect the pattern to predict breast cancer. The proposed work is supposed to implement the proposed methodology. A comparison of the proposed methodology and algorithm with the existing algorithm is made during simulation. The proposed work is found to be more efficient compared to the existing techniques used in breast cancer prediction. The utilization of proposed in the work area of medical science is supposed to enhance the capability in case of CNN at the time of decision-making. The proposed work is supposed to be more accurate compared to the existing works. It has been observed that the proposed work is fourteen to fifteen percent more accurate. It is taking 9/4 times less space and 1.0849004/0.178971 times less time compared to the general CNN model. Accuracy might vary as per size of the image and alteration performed in dataset of the image. [ABSTRACT FROM AUTHOR]