Intelligent detection of marine organism plays an important part in the marine economy, and it is significant to detect marine organisms quickly and accurately in a complex marine environment for the intelligence of marine equipment. The existing object detection models do not work well underwater. This paper improves the structure of EfficientDet detector and proposes the EfficientDet-Revised (EDR), which is a new marine organism object detection model. Specifically, the MBConvBlock is reconstructed by adding the Channel Shuffle module to enable the exchange of information between the channels of the feature layer. The fully connected layer of the attention module is removed and convolution is used to cut down the amount of network parameters. The Enhanced Feature Extraction module is constructed for multi-scale feature fusion to enhance the feature extraction ability of the network to different objects. The results of experiments demonstrate that the mean average precision (mAP) of the proposed method reaches 91.67% and 92.81% on the URPC dataset and the Kaggle dataset, respectively, which is better than other object detection models. At the same time, the processing speed reaches 37.5 frame per second (FPS) on the URPC dataset, which can meet the real-time requirements. It can provide a useful reference for underwater robots to perform tasks such as intelligent grasping.

Recent research has shown that spatial-spectral information can help to improve the classification of hyperspectral images (HSIs). Therefore, three-dimensional convolutional neural networks (3D-CNNs) have been applied to HSI classification. However, a lack of HSI training samples restricts the performance of 3D-CNNs. To solve this problem and improve the classification, an improved method based on 3D-CNNs combined with parameter optimization, transfer learning, and virtual samples is proposed in this paper. Firstly, to optimize the network performance, the parameters of the 3D-CNN of the HSI to be classified (target data) are adjusted according to the single variable principle. Secondly, in order to relieve the problem caused by insufficient samples, the weights in the bottom layers of the parameter-optimized 3D-CNN of the target data can be transferred from another well trained 3D-CNN by a HSI (source data) with enough samples and the same feature space as the target data. Then, some virtual samples can be generated from the original samples of the target data to further alleviate the lack of HSI training samples. Finally, the parameter-optimized 3D-CNN with transfer learning can be trained by the training samples consisting of the virtual and the original samples. Experimental results on real-world hyperspectral satellite images have shown that the proposed method has great potential prospects in HSI classification.

Although the underwater optical communication system (UWOC) plays an important role in many marine applications, the acquisition, pointing and tracking (APT) problems are still great challenges in a long or turbid water channel. The Orthogonal Frequency Division Multiplexing (OFDM) and the space-time block coding (STBC) technologies have been introduced in UWOC systems to improve spectral efficiency and data transmission reliability. On these foundations, a multi-input multi-output (MIMO) mode is proposed and verified in this article to reduce the alignment requirement. In experiments, two sets of 450 nm blue laser diodes (LDs) and PIN photodiodes are selected as light sources and receivers, and then parallel light beams are generated and modulated using an Alamouti coding for its anti-interference ability. The receiving plane is rotated to simulate the various alignment situations, and the system performance of MIMO-OFDM and single-input single-output OFDM (SISO-OFDM) are compared in tap water and turbid water. The results show that the spatial diversity in MIMO can extend the communication range and its single-sided maximum detection angle range is 92% and 112% higher than the SISO system, respectively.

As a straightforward method for human-machine interaction, gesture recognition is vital in many practical applications. However, effectively extracting spatiotemporal information from video is still a fundamental problem and designing an accurate and efficient network is a feasible solution. The ConvNeXt, renowned for its superior still image processing capabilities, is chosen as the basis of this work. Then, the network is extended to a three-dimensions patten for dynamic data and a two-scale convolution kernel is introduced to focus on the hand region. Therefore, a novel Two-scale 3D-ConvNeXt Network (TS3C-Net) is established. Furthermore, the Mixup, Cutmix data augmentation and label smoothing regularization are also applied to enhance the performance further. The experiments show that the accuracy of the proposed TS3C-Net achieves 95.36%, 97.1% and 87.55% on EgoGesture, Jester and NVGesture datasets, respectively.

With the development of remote sensing technology, classification has become a meaningful way to explore the rich information in hyperspectral images (HSIs). However, various environmental factors may cause noise and shadow areas in HSIs, resulting in weak signals and difficulties in fully utilizing information. In addition, classification methods based on deep learning have made considerable progress, but features extracted from most networks have much redundancy. Therefore, a method based on two-dimensional dynamic stochastic resonance (2D DSR) shadow enhancement and convolutional neural network (CNN) classification combined with an attention mechanism (AM) for HSIs is proposed in this paper. Firstly, to protect the spatial correlation of HSIs, an iterative equation of 2D DSR based on the pixel neighborhood relationship was derived, which made it possible to perform matrix SR in the spatial dimension of the image, instead of one-dimensional vector resonance. Secondly, by using the noise in the shadow area to generate resonance, 2D DSR can help increase the signals in the shadow regions by preserving the spatial characteristics, and enhanced HSIs can be obtained. Then, a 3DCNN embedded with two efficient channel attention (ECA) modules and one convolutional block attention module (CBAM) was designed to make the most of critical features that significantly affect the classification accuracy by giving different weights. Finally, the performance of the proposed method was evaluated on a real-world HSI, and comparative studies were carried out. The experimental results showed that the proposed approach has promising prospects in HSIs’ shadow enhancement and information mining.