The effective segmentation of esophagus and esophageal cancer from Computed Tomography (CT) images can meaningfully assist doctors in the diagnosis and treatment of esophageal cancer patients. However, problems such as the small proportion of the esophageal region in CT images and the irregular shape of the esophagus will make the diagnosis difficult. In practical applications, not all esophagus and esophageal cancer morphology can be included in the training set, so the generalization ability of the model is very important. These are the difficulties in segmenting the esophagus and esophageal cancer. Since some adjacent tissues and organs of the esophagus are visually close to the esophagus and esophageal cancer, how to ensure that the network can extract effective distinguishing features has become the focus of research. In this paper, a novel U-Net structure — Channel-attention U-Net is proposed to segment esophagus and esophageal cancer from CT slices. This novel network combines a Channel Attention Module (CAM) that can distinguish the esophagus and surrounding tissues by emphasizing and inhibiting channel feature and Cross-level Feature Fusion Module (CFFM) which is utilized to strengthen the generalization ability of the network by using high-level features to weight low-level features. Because the high-level features represent specific organizational information, and the low-level features represent the characteristics of detailed information such as edges and contours, the network can learn specific detailed features of a definite organization. In addition, to locate the esophageal region better, a 3D semi-automatic method for segmenting esophagus and esophageal cancer is proposed. The proposed network is trained using 46,400 CT pictures as the training set and divides 11,600 CT images from the dataset at a ratio of 0.2 as the validation set. Finally, 7,250 CT images were used as the test set to test the performance of the network. The experimental results show that the IoU value of our network can reach 0.625, the dice value is 0.732 and the Hausdorff distance is 3.193.

Knowledge graph technology has important guiding significance for efficient and orderly fault diagnosis of robot transmission system. Taking the historical robot maintenance logs of robot transmission system as the research object, a top-down fault diagnosis event logic knowledge graph construction method is proposed. Firstly, we define event arguments of fault phenomenon and fault cause events, define event argument classes and relation between classes, and construct an event logic knowledge ontology model. According to the event logic knowledge ontology, the fault diagnosis event argument entity and relation in the corpus are labeled, and an event logic knowledge extraction dataset is formed. Secondly, an event argument entity and relation joint extraction model is proposed. Using stacked bidirectional long short-term memory(BiLSTM) to obtain deep context features of text. As a supplement to stacked BiLSTM, self-attention mechanism extracts character dependency features from multiple subspaces, and uses conditional random field(CRF) to realize entity recognition. The character dependency features are mapped to the entity label weight embedding, and spliced with deep context features to extract relations. Bidirectional graph convolutional network(BiGCN) is introduced for relation inference, graph convolution features are used to update deep context features to perform joint extraction in the second phase. Experimental results show that this method can improve the effect of event argument entity and relation joint extraction and is better than other methods. Finally, an event logic knowledge graph of robot transmission system fault diagnosis is constructed, which provides decision support for autonomous fault diagnosis of robot transmission system.

Cross-domain few-shot learning (FSL) often employs adversarial domain adaptation techniques to address the issue of data distribution discrepancies between the source and target domains. However, forcing the alignment of two distinct domains may lead to distortions in class distribution alignment and result in a decrease in classification performance in hyperspectral image analysis. Moreover, existing cross-domain methods are often applied to satellite/airborne hyperspectral image as both the source and target domain. It is rarely explored whether the same cross-domain methods can be applied for cross applications where the source domain and target domain data could be both satellite/airborne hyperspectral image with lower spatial resolution and unmanned aerial vehicle (UAV) hyperspectral image with higher spatial resolution. To address these issues, this paper proposes a novel domain-adaptive FSL network with dual branches respectively aiming at domain fusion and domain separation. The domain fusion branch uses a conditional adversarial network to align the global distributions of the two domains, while the domain separation branch introduces gate mechanism for discriminative feature learning in each domain to achieve independent category distributions. During the experiment, the proposed method is evaluated by performing cross-transfer learning under the condition that low spatial resolution hyperspectral data and high spatial resolution hyperspectral data are used as source and target data alternately. The experimental results suggest that the proposed method not only mitigates the negative effects of forced alignment in domain fusion but also holds potential for cross-domain transfer learning between low and high spatial resolution hyperspectral images.

Due to the complexity of underwater environments and the lack of training samples, the application of target detection algorithms to the underwater environment has yet to provide satisfactory results. It is crucial to design specialized underwater target recognition algorithms for different underwater tasks. In order to achieve this goal, we created a dataset of freshwater fish captured from multiple angles and lighting conditions, aiming to improve underwater target detection of freshwater fish in natural environments. We propose a method suitable for underwater target detection, called DyFish-DETR (Dynamic Fish Detection with Transformers). In DyFish-DETR, we propose a DyFishNet (Dynamic Fish Net) to better extract fish body texture features. A Slim Hybrid Encoder is designed to fuse fish body feature information. The results of ablation experiments show that DyFishNet can effectively improve the mean Average Precision (mAP) of model detection. The Slim Hybrid Encoder can effectively improve Frame Per Second (FPS). Both DyFishNet and the Slim Hybrid Encoder can reduce model parameters and Floating Point Operations (FLOPs). In our proposed freshwater fish dataset, DyFish-DETR achieved a mAP of 96.6%. The benchmarking experimental results show that the Average Precision (AP) and Average Recall (AR) of DyFish-DETR are higher than several state-of-the-art methods. Additionally, DyFish-DETR, respectively, achieved 99%, 98.8%, and 83.2% mAP in other underwater datasets.

The application of computer vision in fish identification facilitates researchers and managers to better comprehend and safeguard the aquatic ecological environment. Numerous researchers have harnessed deep learning methodologies for studying fish species identification. Nonetheless, this endeavor still encounters challenges such as high computational costs, a substantial number of parameters, and limited practicality. To address these issues, we propose a lightweight network architecture incorporating deformable convolutions, termed DeformableFishNet. Within DeformableFishNet, an efficient global coordinate attention module (EGCA) is introduced alongside a deformable convolution network (EDCN/EC2f), which is grounded in EGCA, to tackle the deformation of fish bodies induced by swimming motions. Additionally, an EC2f-based feature pyramid network (EDBFPN) and an efficient multi-scale decoupling head (EMSD Head) are proposed to extract multi-scale fish features within a lightweight framework. DeformableFishNet was deployed on our freshwater fish dataset, with experimental outcomes illustrating its efficacy, achieving a mean average precision (mAP) of 96.3%. The model comprises 1.7 million parameters and entails 4.7 billion floating-point operations (FLOPs). Furthermore, we validated DeformableFishNet on three public underwater datasets, yielding respective mAPs of 98%, 99.4%, and 83.6%. The experiments show that DeformableFishNet is suitable for underwater identification of various scenes.

Monitoring crop growth is of great significance to obtain crop growth status information for development of smart agriculture. The traditional way to measure the phenotypic parameters of crops is labor-intensive and encounters inconvenient operations. In this study, we propose to obtain the phenotypic parameters of crops from 3-D reconstruction of plants from single RGB images using a data-driven plant phenotypic parameters estimation network (P3ES-Net) deep neural network, which enables to estimate the depth shift and camera focal length used for depth estimation and reconstruction of the 3-D model of plants. Based on the principles of the monocular ranging and pinhole imaging model, crop phenotypic parameters such as height, canopy size, and trunk diameter can then be calculated from the 3-D model. Experiments with four practical plants present that our method is able to achieve acceptable evaluation of the growth status of plants. Of more significance, it achieves particular superior depth estimation performance over a commercial depth camera, which is a very new on-sale depth camera using stereo vision and deep learning network. This potential performance throws light on the low-cost measurement of crop phenotypic parameters using RGB camera in monitoring crop growth.