Automatic defect detection on the metal surface is a vital task for product inspection in industrial assembly lines or production processes. Owing to miscellaneous patterns of defects, interclass similarity, intraclas...
详细信息
Automatic defect detection on the metal surface is a vital task for product inspection in industrial assembly lines or production processes. Owing to miscellaneous patterns of defects, interclass similarity, intraclass difference, and fewer defect samples, achieving accurate and automatic detection remains a big challenge. What is more, since the rising demand for production efficiency, real-time detection is increasingly desirable. This article proposes a semantic prior and extremely efficient dilated convolution network, named SPEED, for pixel-wise detection on the metal surface, which aims to address the aforementioned issues. The architecture of SPEED involves the following: 1) a semantic prior (SP) branch, with shallow layer and prior mapping module to capture low-level details;and 2) an extremely efficient dilation (EED) branch, with lightweight bottleneck to obtain high-level context. Furthermore, an aggregation module is designed to fuse both types of feature representation. Additionally, different level features of bottleneck are fused to improve the segmentation performance. Experimental results on three metal surface defect datasets indicate that the proposed method outperforms the state-of-the-art approaches in terms of the mean intersection of union, model parameters, FLOPs, and FPS. More specifically, SPEED achieves 92.34% mIoU on NEU-Seg, 88.65% mIoU on Severstal Strip Steel, and 63.91% mIoU on MT Defect.
Implicit scene completion aims to learn an implicit representation of dense point clouds from incomplete ones. Since point clouds are disordered and irregular, some implicit scene completion methods learn representati...
详细信息
ISBN:
(数字)9798350390155
ISBN:
(纸本)9798350390162
Implicit scene completion aims to learn an implicit representation of dense point clouds from incomplete ones. Since point clouds are disordered and irregular, some implicit scene completion methods learn representations from voxelized point clouds with sparse convolution. Despite achieving promising results, they lack deep exploration of feature learning on empty voxels, which is beneficial for implicit scene completion task. To address this, we propose a dense voxel representation network for implicit scene completion. First, we design a Bird’s-Eye View (BEV) assisted enhancement module to enhance non-empty voxel features by incorporating the information contained in the learned dense BEV features into them through deformable cross-attention. Second, we construct a feature adaptive completion module to adaptively complete voxel features using deformable self-attention, realizing the transfer of the information from non-empty voxels to empty voxels. Extensive experiments on SemanticKITTI and SemanticPOSS datasets demonstrate our method achieves state-of-the-art performance.
The accumulation of time-series signals and the absence of labels make time-series Anomaly Detection (AD) a self-supervised task of deep learning. Methods based on normality assumptions face the following three limita...
详细信息
To address the problems of inadequate feature interaction and lack of targeting in feature combination in the click-through rate prediction model. We propose a click-through prediction model called SELFM. It based on ...
详细信息
Additive manufacturing(AM)technology enables the creation of a wide variety of assemblies and complex shapes from three-dimensional model data in a bottom-up,layer-by-layer ***,AM has revolutionized the modern manufac...
详细信息
Additive manufacturing(AM)technology enables the creation of a wide variety of assemblies and complex shapes from three-dimensional model data in a bottom-up,layer-by-layer ***,AM has revolutionized the modern manufacturing industry,attracting increasing interest from both academic and industrial *** Rapid Manufacturing Center(RMC)of the School of Materials science and Engineering at the Huazhong Univer-sity of science and Technology(HUST),one of the earliest and most powerful AM research teams in China,has been engaged in AM research since *** to address the“stuck neck”problems of specific high-strength products for AM,the RMC has conducted full-chain research in the aspects of special materials,processes,equip-ment,and applications for ***,it has formed a multi-disciplinary research team over the past three *** research achievements in the AM field include winning five national awards,more than ten first prizes,and more than ten second prizes at the provincial and ministerial *** RMC was complimented as“the world’s most influential organization in the laser AM field in 2018”by Virtual and Physical Prototyping(an international authoritative magazine of AM).Moreover,their industrialization achievements were evaluated as“having affected countries such as Singapore,South Korea,and the United States”by an international author-itative Wohlers Report on *** this study,we first summarize the representative research achievements of the RMC in the AM *** include the preparation and processing technology of high-performance polymeric,metallic,and ceramic materials for AM;advanced processing technology and software/equipment for AM;and typical AM-fabricated products and their ***,we discuss the latest research achievements in cutting-edge 4D printing in terms of feedstock selection,printing processes,induction strategies,and potential ***,we provide insights into the future di
During online social networks (OSNs), popularity prediction uncovers the final size of online content based on the observed cascade, which has been the critical technology for online recommendation, viral marketing, a...
详细信息
During online social networks (OSNs), popularity prediction uncovers the final size of online content based on the observed cascade, which has been the critical technology for online recommendation, viral marketing, and rumor detection. Recently, representation learning could help to infer the mapping between the dynamic cascade and the final popularity efficiently, and has been a new research paradigm for popularity prediction. However, those methods are vulnerable to structure disturbance when lack of fine-grained supervision, as only the dynamic cascade is used. Therefore, we propose a novel trend and cascade based spatiotemporal evolution network (TCSE-Net), which preserves the distinguishable structure pattern while eliminating potential noise, via aligning and fusing the temporal popularity and cascade. To be specific, we first leveraged the Long-Short Term Memory (LSTM) and recurrent graph convolutional network (GCN) to learn the trend representation and the corresponding cascade representation respectively. Meanwhile, we represent node with it's layer, thereby the hierarchy is preserved in cascade representation through GCN. Then, both trend and cascade representations are aligned in time sequence and selectively assembled by a set of shared parameters for popularity prediction. The extensive experimental results show that our TCSE-Net outperforms state-of-the-art baselines on two real datasets. Related code will be publicly available on https://***/TAN-OpenLab/TCSE-Net.
To minimize the length of scheduling and guarantee the load balance of channels, a Load-balanced and length-minimized link scheduling (LBLM) algorithm is proposed. LBLM algorithm is a heuristic scheme, which assigns t...
详细信息
To minimize the length of scheduling and guarantee the load balance of channels, a Load-balanced and length-minimized link scheduling (LBLM) algorithm is proposed. LBLM algorithm is a heuristic scheme, which assigns time slots for unicast traffic based on link's weight and hop-count in the routing traffic tree. Thus the algorithm considers both primary and secondary interference, as well as guarantees the proportional fairness. The ns2 simulation results show that in multi-channel TDMA Wireless mesh networks (WMNs), the proposed algorithm has the benefits of lower complexity, shorter frame length and better channel balance compared to other well-known schedule mechanisms.
Artificial intelligence techniques are used in many areas today to find solutions to different problems. Scientists are trying to solve some problems in people’s daily lives using these techniques. To solve these pro...
详细信息
Cultural heritage institutions are exploring Semantic Web technologies to publish and enrich their catalogues. Several initiatives, such as Labs, are based on the creative and innovative reuse of the materials publish...
详细信息
Cultural heritage institutions are exploring Semantic Web technologies to publish and enrich their catalogues. Several initiatives, such as Labs, are based on the creative and innovative reuse of the materials published by cultural heritage institutions. In this way, quality has become a crucial aspect to identify and reuse a dataset for research. In this article, we propose a methodology to create Shape Expressions definitions in order to validate LOD datasets published by libraries. The methodology was then applied to four use cases based on datasets published by relevant institutions. It intends to encourage institutions to use ShEx to validate LOD datasets as well as to promote the reuse of LOD, made openly available by libraries.
Recently, gigapixel photography has been developed considerably and gradually put into remote sensing, video surveillance, etc. Gigapixel images have a visible field of view area at the square-kilometer level (contain...
详细信息
Recently, gigapixel photography has been developed considerably and gradually put into remote sensing, video surveillance, etc. Gigapixel images have a visible field of view area at the square-kilometer level (containing thousands of targets) and up to 100 times the scale variation. Among them, the differences in target pose, scale, and occlusion are huge, and most existing target detection algorithms cannot directly process them. To solve these problems, we propose a new multi-target pedestrian and vehicle detector PVDet (Towards Pedestrian and Vehicle Detection on Gigapixel-level images) for gigapixel-level images. First, the DPRNet (Deformable deeP Residual Network) is designed as the backbone network to enhance the effective perceptual field and improve the feature representation of pose varying and occluded targets. Then, the PAFPN (Path Aggregation Feature Pyramid Network) is adopted to process the multi-scale features extracted by the backbone, boosting the multi-scale target modeling capability and the localization of small targets. Finally, the DyHead module is introduced to enhance the detection head's scale, spatial and task awareness, further optimizing pedestrian and vehicle classification and localization. Compared with other State-of-the-Art methods on the PANDA dataset, the experimental results show that the proposed method dramatically improves AP of pedestrian and vehicle detection in gigapixel-level images by 10.4 AP over baseline, which is better than the existing target detection algorithms. We also conducted experiments on the PASCAL VOC 2012 dataset to further demonstrate the generalization capability and effectiveness of the proposed method.
暂无评论