Surface defect detection of hot rolled steel based on multi-scale feature fusion and attention mechanism residual block.

To improve the precision of defect categorization and localization in images, this paper proposes an approach for detecting surface defects in hot-rolled steel strips. The approach uses an improved YOLOv5 network model to overcome the issues of inadequate feature extraction capacity and suboptimal feature integration when identifying surface defects on steel strips. The proposed method achieves higher detection accuracy and localization precision, making it more competitive and applicable in real production. Firstly, the multi-scale feature fusion (MSF) strategy is utilized to fuse shallow and deep features effectively and enrich detailed information relevant to target defects. Secondly, the CSPLayer Res2Attention block (CRA block) residual module is introduced to reduce the loss of defect information during hierarchical transmission, thereby enhancing the extraction of fine-grained features and improving the perception of details and global features. Finally, the experimental results indicate that the mAP on the NEU-DET and GC10-DET datasets approaches 78.5% and 67.3%, respectively, which is 4.9% and 2.1% higher than that of the baseline. Meanwhile, it has higher precision and more precise localization capabilities than other methods. Furthermore, it also achieves 59.2% mAP on the APDDD dataset, indicating its potential for growth in further domains.

Cross-domain mechanism for few-shot object detection on Urine Sediment Image.

Deep learning object detection networks require a large amount of box annotation data for training, which is difficult to obtain in the medical image field. The few-shot object detection algorithm is significant for an unseen category, which can be identified and localized with a few labeled data. For medical image datasets, the image style and target features are incredibly different from the knowledge obtained from training on the original dataset. We propose a background suppression attention(BSA) and feature space fine-tuning module (FSF) for this cross-domain situation where there is a large gap between the source and target domains. The background suppression attention reduces the influence of background information in the training process. The feature space fine-tuning module adjusts the feature distribution of the interest features, which helps to make better predictions. Our approach improves detection performance by using only the information extracted from the model without maintaining additional information, which is convenient and can be easily plugged into other networks. We evaluate the detection performance in the in-domain situation and cross-domain situation. In-domain experiments on the VOC and COCO datasets and the cross-domain experiments on the VOC to medical image dataset UriSed2K show that our proposed method effectively improves the few-shot detection performance.

Multi-Target Tracking Based on a Combined Attention Mechanism and Occlusion Sensing in a Behavior-Analysis System.

Multi-object tracking (MOT) is a topic of great interest in the field of computer vision, which is essential in smart behavior-analysis systems for healthcare, such as human-flow monitoring, crime analysis, and behavior warnings. Most MOT methods achieve stability by combining object-detection and re-identification networks. However, MOT requires high efficiency and accuracy in complex environments with occlusions and interference. This often increases the algorithm's complexity, affects the speed of tracking calculations, and reduces real-time performance. In this paper, we present an improved MOT method combining an attention mechanism and occlusion sensing as a solution. A convolutional block attention module (CBAM) calculates the weights of space and channel attention from the feature map. The attention weights are used to fuse the feature maps to extract adaptively robust object representations. An occlusion-sensing module detects an object's occlusion, and the appearance characteristics of an occluded object are not updated. This can enhance the model's ability to extract object features and improve appearance feature pollution caused by the short-term occlusion of an object. Experiments on public datasets demonstrate the competitive performance of the proposed method compared with the state-of-the-art MOT methods. The experimental results show that our method has powerful data association capability, e.g., 73.2% MOTA and 73.9% IDF1 on the MOT17 dataset.

Investigation on the Wilson Neuronal Model: Optimized Approximation and Digital Multiplierless Implementation.

Neuromorphic engineering is an essential science field which incorporates the basic aspects of issues together such as: physics, mathematics, electronics, etc. The primary block in the Central Nervous System (CNS) is neurons that have functional roles such as: receiving, processing, and transmitting data in the brain. This paper presents Wilson Multiplierless Neuron (WMN) model which is a modified version of the original model. This model uses power-2 based functions, Look-Up Table (LUT) approach and shifters to apply a multiplierless digital realization leads to overhead costs reduction and increases in the final system frequency. The proposed model specifically follows the original neuron model in case of spiking patterns and also dynamical pathways. To validate the proposed model in digital hardware implementation, the FPGA board (Xilinx Virtex II XC2VP30) can be used. Hardware results show the increasing in the system frequency compared with the original model and other similar papers. Numerical results demonstrate that the proposed system speed-up is 210 MHz that is higher than the original one, 85 MHz. Additionally, the overall saving in FPGA resources for the proposed model is 96.86 % that is more than the original model, 95.13 %. From case study viewpoint for CNS consideration, a network consisting of Wilson neurons, synapses, and astrocytes have been considered to test the controlling effects on LTP and LTD processes for investigating the neuronal diseases (medical approaches) such as Epilepsy.

Siamese Implicit Region Proposal Network With Compound Attention for Visual Tracking.

Recently, siamese-based trackers have achieved significant successes. However, those trackers are restricted by the difficulty of learning consistent feature representation with the object. To address the above challenge, this paper proposes a novel siamese implicit region proposal network with compound attention for visual tracking. First, an implicit region proposal (IRP) module is designed by combining a novel pixel-wise correlation method. This module can aggregate feature information of different regions that are similar to the pre-defined anchor boxes in Region Proposal Network. To this end, the adaptive feature receptive fields then can be obtained by linear fusion of features from different regions. Second, a compound attention module including a channel and non-local attention is raised to assist the IRP module to perform a better perception of the scale and shape of the object. The channel attention is applied for mining the discriminative information of the object to handle the background clutters of the template, while non-local attention is trained to aggregate the contextual information to learn the semantic range of the object. Finally, experimental results demonstrate that the proposed tracker achieves state-of-the-art performance on six challenging benchmark tests, including VOT-2018, VOT-2019, OTB-100, GOT-10k, LaSOT, and TrackingNet. Further, our obtained results demonstrate that the proposed approach can be run at an average speed of 72 FPS in real time.

Active contours driven by edge entropy fitting energy for image segmentation.

Active contour models have been widely used for image segmentation purposes. However, they may fail to delineate objects of interest depicted on images with intensity inhomogeneity. To resolve this issue, a novel image feature, termed as local edge entropy, is proposed in this study to reduce the negative impact of inhomogeneity on image segmentation. An active contour model is developed on the basis of this feature, where an edge entropy fitting (EEF) energy is defined with the combination of a redesigned regularization term. Minimizing the energy in a variational level set formulation can successfully drive the motion of an initial contour curve towards optimal object boundaries. Experiments on a number of test images demonstrate that the proposed model has the capability of handling intensity inhomogeneity with reasonable segmentation accuracy.