A complete defect detection task aims to achieve the specific class and precise location of each defect in an image, which makes it still challenging for applying this task in practice. The defect detection is a composite task of classification and location, leading to related methods are often hard to take into account the accuracy of both. And the implementation of defect detection depends on a special detection dataset which contains expensive manual annotations. In this paper, we proposed a novel defect detection system based on deep learning and focused on a practical industrial application: steel plate defect inspection. In order to achieve strong classification-ability, this system employs a baseline convolution neural network (CNN) to generate feature maps at each stage. And then the proposed multilevel-feature fusion network (MFN) combines multiple hierarchical features into one feature, which can include more location details of defects. Based on these multilevel features, a region proposal network (RPN) is adopted to generate regions of interest (ROIs). For each ROI, a detector, consisting of a classifier and a bounding box regressor, produces the final detection results. Finally, we set up a defect detection dataset NEU-DET for training and evaluating our method. On the NEU-DET, our method achieves 74.8/82.3 mAP with baseline networks ResNet34/50 by using 300 proposals. In addition, by using only 50 proposals, our method can detect at 20 fps on a single GPU and reach 92% of the above performance, hence the potential for real-time detection.

Surface defect detection is a critical task in industrial production process. Nowadays, there are lots of detection methods based on computer vision and have been successfully applied in industry, they also achieved good results. However, achieving full automation of surface defect detection remains a challenge, due to the complexity of surface defect, in intra-class, while the defects between inter-class contain similar parts, there are large differences in appearance of the defects. To address these issues, this paper proposes a pyramid feature fusion and global context attention network for pixel-wise detection of surface defect, called PGA-Net. In the framework, the multi-scale features are extracted at first from backbone network. Then the pyramid feature fusion module is used to fuse these features into five resolutions through some efficient dense skip connections. Finally, the global context attention module is applied to the fusion feature maps of adjacent resolution, which allows effective information propagate from low-resolution fusion feature maps to high-resolution fusion ones. In addition, the boundary refinement block is added to the framework to refine the boundary of defect and improve the result of predict. The final prediction is the fusion of the five resolutions fusion feature maps. The results of evaluation on four real-world defect datasets demonstrate that the proposed method outperforms the state-of-the-art methods on mean Intersection of Union and mean Pixel Accuracy (NEU-Seg: 82.15%, DAGM 2007: 74.78%, MT_defect: 71.31%, Road_defect: 79.54%).

Visual perception refers to the process of organizing, identifying and interpreting visual information in environmental awareness and understanding. With the rapid progress of multimedia acquisition technology, research on visual perception has been a hot topic in the academical field and industrial applications. Especially after the introduction of artificial intelligence theory, intelligent visual perception has been widely used to promote the development of industrial production towards intelligence. In this paper, we review the previous research and application of visual perception in different industrial fields such as product surface defect detection, intelligent agricultural production, image synthesis, event reconstruction, intelligent driving and pose measurement. The applications basically cover most of the intelligent visual perception processing technologies. Through this survey, it will provide a comprehensive reference for research on this direction. Finally, this paper also summarizes the current challenges of visual perception and predicts its future development trends.

This paper considers the problem of maximizing the number of task allocations in a distributed multirobot system under strict time constraints, where other optimization objectives need also be considered. It builds upon existing distributed task allocation algorithms, extending them with a novel method for maximizing the number of task assignments. The fundamental idea is that a task assignment to a robot has a high cost if its reassignment to another robot creates a feasible time slot for unallocated tasks. Multiple reassignments among networked robots may be required to create a feasible time slot and an upper limit to this number of reassignments can be adjusted according to performance requirements. A simulated rescue scenario with task deadlines and fuel limits is used to demonstrate the performance of the proposed method compared with existing methods, the consensus-based bundle algorithm and the performance impact (PI) algorithm. Starting from existing (PI-generated) solutions, results show up to a 20% increase in task allocations using the proposed method.

Visual information has been paid more attention to rail surface defect detection for its high efficiency and stability. However, it is not sufficient to detect more complete defects in complex background information. The addition of profiles can effectively improve the above situation due to its entity information. However, in high-speed detection, traditional three-dimensional (3D) profiles acquisition is difficult and separate from image acquisition, which can not satisfy the above requirements effectively. Therefore, an unsupervised stereoscopic saliency detection method based on binocular line-scanning system is proposed in this paper. This method can obtain the extraordinary precision image and profile information at the same time, and avoid decoding distortion of the structured light reconstruction method. In this method, a global low-rank non-negative reconstruction algorithm with a background constraint is proposed. Unlike the low-rank recovery (LRR) model, the algorithm has a more comprehensive low-rank and background clustering properties. Besides, outlier detection based on the geometric properties of the rail surface is also proposed in this method. Finally, image saliency results and depth outlier detection results are associated with interactive fusion. Otherwise, data set (RSDDS-113) containing rail surface defects is established for experimental verification. The experimental results demonstrate that our method can obtain the results that MAE is 0.09 and AUC is 0.94, which is better than other 15 algorithms.

The trustworthiness of robots and autonomous systems (RAS) has taken a prominent position on the way towards full autonomy. This work is the first to systematically explore the key facets of human-centred AI for trustworthy RAS. We identified five key properties of a trustworthy RAS, i.e., RAS must be (i) safe in any uncertain and dynamic environment; (ii) secure, i.e., protect itself from cyber threats; (iii) healthy and fault-tolerant; (iv) trusted and easy to use to enable effective human-machine interaction (HMI); (v) compliant with the law and ethical expectations. While the applications of RAS have mainly focused on performance and productivity, not enough scientific attention has been paid to the risks posed by advanced AI in RAS. We analytically examine the challenges of implementing trustworthy RAS with respect to the five key properties and explore the role and roadmap of AI technologies in ensuring the trustworthiness of RAS in respect of safety, security, health, HMI, and ethics. A new acceptance model of RAS is provided as a framework for human-centric AI requirements and for implementing trustworthy RAS by design. This approach promotes human-level intelligence to augment human capabilities and focuses on contribution to humanity.