Abstract Existing deep learning methods cannot achieve satisfactory ship license plate (SLP) recognition due to the harsh marine environment, such as foggy weather, unstable ship state and small targets. Therefore, a convolutional recurrent neural network (CRNN)‐based method is proposed for accurate SLP image recognition. Overall, the suggested method improves a CRNN recognition model by SLP image enhancement and data augmentation. The SLP image enhancement employs dark channel prior and Hough transform line detector to address the fog/blurriness and tilt issues existing in SLP images. As separate and joint operations, the two enhancements contribute to data augmentation for CRNN recognition. Preprocessing algorithms, including adaptive histogram equalization and image edge padding, are used to improve and unify the enlarged dataset for augmenting the CRNN model. As a final step, correction of the CRNN recognition results is made according to the character rule of SLPs, using an edit‐distance algorithm to match against a pre‐established SLP dictionary. A variety of real SLP images were collected to build an SLP image dataset for verification. The experimental results indicate that our method can reach an SLP recognition accuracy of , which is significantly superior to other text‐based deep learning methods.

Aiming to obtain more discriminative features in scene images and overcome the impacts of intra-class differences and inter-class similarities, the paper proposes a scene recognition method that combines attention and context information. First, we introduce the attention mechanism and build a multi-scale attention model. Discriminative information considers salient objects and regions by means of channel attention and spatial attention. Besides, the central loss function joint supervision strategy is introduced to further reduce the misjudgment of intra-class differences. Second, a model based on multi-level context information is proposed to describe the positional relationship between objects, which can effectively alleviate the influence of the similarity of objects between classes. Finally, the two models are merged to give full play to the compatibility of features, so that the final feature representation not only focuses on the effective discriminant information, but also manifests the relative position relationship between significant objects. Extensive experiments have proved that the method in this paper effectively solves the problem of insufficient feature representation in scene recognition tasks, and improves the accuracy of scene recognition.

Gesture is a natural form of human communication, and it is of great significance in human–computer interaction. In the dynamic gesture recognition method based on deep learning, the key is to obtain comprehensive gesture feature information. Aiming at the problem of inadequate extraction of spatiotemporal features or loss of feature information in current dynamic gesture recognition, a new gesture recognition architecture is proposed, which combines feature fusion network with variant convolutional long short‐term memory (ConvLSTM). The architecture extracts spatiotemporal feature information from local, global and deep aspects, and combines feature fusion to alleviate the loss of feature information. Firstly, local spatiotemporal feature information is extracted from video sequence by 3D residual network based on channel feature fusion. Then the authors use the variant ConvLSTM to learn the global spatiotemporal information of dynamic gesture, and introduce the attention mechanism to change the gate structure of ConvLSTM. Finally, a multi‐feature fusion depthwise separable network is used to learn higher‐level features including depth feature information. The proposed approach obtains very competitive performance on the Jester dataset with the classification accuracies of 95.59%, achieving state‐of‐the‐art performance with 99.65% accuracy on the SKIG (Sheffifield Kinect Gesture) dataset.