The automatic defects detection for solar cell electroluminescence (EL) images is a challenging task, due to the similarity of defect features and complex background features. To address this problem, a novel Complementary Attention Network (CAN) is designed by connecting the novel channel-wise attention subnetwork with spatial attention subnetwork sequentially, which adaptively suppresses the background noise features and highlights the defect features simultaneously by employing the complementary advantage of the channel features and spatial position features. In CAN, the novel channel-wise attention subnetwork applies convolution operation to integrate the concatenated and discriminative output features extracted by Global Average Pooling (GAP) layer and Global Max Pooling (GMP) layer, which can make fully use of these informative features. Furthermore, a Region Proposal Attention Network (RPAN) is proposed by embedding CAN into Region Proposal Network (RPN) in Faster R-CNN to extract more refined defective region proposals, which is used to construct a novel end-to-end Faster RPAN-CNN framework for detecting defects in raw EL image. Finally, some experimental results on a large-scale EL dataset including 3629 images, 2129 of which are defective, show that the proposed method performs much better than other methods in terms of defects classification and detection results in raw solar cell EL images.

The small hot spot defect detection for photovoltaic (PV) farms is a challenging problem due to the feature vanishing as the network deepens. To solve this challenging problem, a novel residual channelwise attention gate network (RCAG-Net) is proposed by employing a novel RCAG module to achieve multiscale feature fusion, complex background suppression, and defect feature highlighting. In RCAG-Net, the novel RCAG module first realizes feature fusion by adding the features of different scale layers. Next, global average pooling (GAP) and multilayer perceptron (MLP) are used to dimension reduction and refinement of the fused features, then yielding an attention map for channelwise feature reweighting by gate mechanism, which employs selective transmission of the convolution neural network (CNN)-extracted features to achieve informative feature filtering. Moreover, residual connection from the fused features to the final output facilitates the insertion of the new RCAG into some classical pretrained models, without breaking its initial behavior. Finally, the proposed approach is validated through a real defect detection system, and the experimental result clearly verifies its effectiveness for small hot spot detection of PV farms.

For a crane system, when the payload is too large to be seen as a mass point, or the hook mass cannot be directly ignored, it performs more like a double-pendulum crane system, instead of a single-pendulum crane system. Due to many factors, the working environment of the industrial crane system is complex. Obstacles sometimes appear in the movement path of the payload, which affects the normal operation of the crane system and may cause accidents like collisions. To handle this issue, we propose a time optimal trajectory planning method for the double-pendulum crane system with obstacle avoidance, which ensures the objective of obstacle avoidance by the payload hoisting/lowering. During the trajectory planning process, a series of physical constraints, including the trolley velocity constraints, the trolley acceleration constraints, the payload’s swing angle constraints and the hook’s swing angle constraints are considered. It can improve the safety of the crane system during the entire working process and the transportation efficiency is also improved at the same time. Finally, the effectiveness of the proposed method is verified by simulations.

The Internet of Things (IoT) technology has been widely introduced to the existing medical system. An eHealth system based on IoT devices has gained widespread popularity. In this article, we propose an IoT eHealth framework to provide an autonomous solution for patients with interventional cardiovascular diseases. In this framework, wearable sensors are used to collect a patient's health data, which is daily monitored by a remote doctor. When the monitoring data is abnormal, the remote doctor will ask for image acquisition of the patient's cardiovascular internal conditions. We leverage edge computing to classify these training images by the local base classifier; thereafter, pseudo-labels are generated according to its output. Moreover, a deep segmentation network is leveraged for the segmentation of stent structs in intravascular optical coherence tomography and intravenous ultrasound images of patients. The experimental results demonstrate that remote and local doctors perform real-time visual communication to complete telesurgery. In the experiments, we adopt the U-net backbone with a pretrained SeResNet34 as the encoder to segment the stent structs. Meanwhile, a series of comparative experiments have been conducted to demonstrate the effectiveness of our method based on accuracy, sensitivity, Jaccard, and dice.