No abstract

No abstract

Traffic sign recognition plays an important role in intelligent transportation systems. Motivated by the recent success of deep learning in the application of traffic sign recognition, we present a shallow network architecture based on convolutional neural networks (CNNs). The network consists of only three convolutional layers for feature extraction, and it learns in a backward optimization way. We propose the method of combining different pooling operations to improve sign recognition performance. In view of real-time performance, we use the activation function ReLU to improve computational efficiency. In addition, a linear layer with softmax-loss is taken as the classifier. We use the German traffic sign recognition benchmark (GTSRB) to evaluate the network on CPU, without expensive GPU acceleration hardware, under real-world recognition conditions. The experiment results indicate that the proposed method is effective and fast, and it achieves the highest recognition rate compared with other state-of-the-art algorithms.

No abstract

This letter presents a new greedy method, called Adaptive Sparsity Matching Pursuit (ASMP), for sparse solutions of underdetermined systems with a typical/random projection matrix. Unlike anterior greedy algorithms, ASMP can extract information on sparsity of the target signal adaptively with a well-designed stagewise approach. Moreover, it takes advantage of backtracking to refine the chosen supports and the current approximation in the process. With these improvements, ASMP provides even more attractive results than the state-of-the-art greedy algorithm CoSaMP without prior knowledge of the sparsity level. Experiments validate the proposed algorithm works well for both noiseless signals and noisy signals, with the recovery quality often outperforming that of -minimization and other greedy algorithms.Index Terms-Adaptive greedy algorithm, blind sparse reconstruction, compressive sensing.

Algorithm frameworks based on feature point matching are mature and widely used in simultaneous localization and mapping (SLAM). However, in the complex and changeable indoor environment, feature point matching-based SLAM currently has two major problems, namely, decreased accuracy of pose estimation due to the interference caused by dynamic objects to the SLAM system and tracking loss caused by the lack of feature points in weak texture scenes. To address these problems, herein, we present a robust and real-time RGB-D SLAM algorithm that is based on ORBSLAM3. For interference caused by indoor moving objects, we add the improved lightweight object detection network YOLOv4-tiny to detect dynamic regions, and the dynamic features in the dynamic area are then eliminated in the algorithm tracking stage. In the case of indoor weak texture scenes, while extracting point features the system extracts surface features at the same time. The framework fuses point and surface features to track camera pose. Experiments on the public TUM RGB-D data sets show that compared with the ORB-SLAM3 algorithm in highly dynamic scenes, the root mean square error (RMSE) of the absolute path error of the proposed algorithm improved by an average of 94.08%. Camera pose is tracked without loss over time. The algorithm takes an average of 34 ms to track each frame of the picture just with a CPU, which is suitably real-time and practical. The proposed algorithm is compared with other similar algorithms, and it exhibits excellent real-time performance and accuracy. We also used a Kinect camera to evaluate our algorithm in complex indoor environment, and also showed high robustness and real-time. To sum up, our algorithm can not only deal with the interference caused by dynamic objects to the system but also stably run in the open indoor weak texture scene.

No abstract

No abstract