KGKPD: A Road Object Detection Algorithm that Fuses Knowledge Graphs and Keypoint Detection

Information on the Internet is fragmented and presented in different data sources, which makes automatic knowledge harvesting and understanding formidable for machines, and even for humans. Knowledge graphs have become prevalent in both of industry and academic circles these years, to be one of the most efficient and effective knowledge integration approaches. Techniques for knowledge graph construction can mine information from either structured, semi-structured, or even unstructured data sources, and finally integrate the information into knowledge, represented in a graph. Furthermore, knowledge graph is able to organize information in an easy-to-maintain, easy-to-understand and easy-to-use manner. In this paper, we give a summarization of techniques for constructing knowledge graphs. We review the existing knowledge graph systems developed by both academia and industry. We discuss in detail about the process of building knowledge graphs, and survey state-of-the-art techniques for automatic knowledge graph checking and expansion via logical inferring and reasoning. We also review the issues of graph data management by introducing the knowledge data models and graph databases, especially from a NoSQL point of view. Finally, we overview current knowledge graph systems and discuss the future research directions.

The knowledge graph, as a significant branch of Library and Information Science, has also become a vital part of Computer Science. In a bid to reveal the research trend and hot topics of the knowledge graph domain, this paper collected latest data about literature on knowledge graphs from SCI-Expanded and EI Compendex database. This paper took advantage of Citespace for visualization analyses of data, which discerned the research topics, hot issues and theme changes of the knowledge graph domain over time. Through analyses, we conclude that the knowledge graph has developed greatly in both Library and Information Science and Computer Science. In Library and Information Science domain, researches focus on the issues like the plotting of knowledge graphs and visualization of knowledge. In Computer Science domain, hot issues are the techniques of constructing knowledge graphs, which include entity recognition, knowledge integration, etc., and their applications in Semantic Web. Among all the research topics, the domain knowledge graph is one of the paramount research trends. Compared with previous researches, this paper keeps up with the times and can reveal the theme evolution of researches on knowledge graphs more precisely.

The micromotion feature of space target provides an effective approach for target recognition. The existing micromotion feature extraction is implemented after target detection and tracking; thus the radar resources need to be allocated for target detection, tracking, and feature extraction, successively. If the feature extraction can be implemented by utilizing the target detecting and tracking pulses, the radar efficiency can be improved. In this paper, by establishing a feedback loop between micromotion feature extraction and track-before-detect (TBD) of target, a novel feature extraction method for space target is proposed. The TBD technology is utilized to obtain the range-slow-time curves of target scatterers. Then, micromotion feature parameters are estimated from the acquired curve information. In return, the state transition set of TBD is updated adaptively according to these extracted feature parameters. As a result, the micromotion feature parameters of space target can be extracted concurrently with implementing the target detecting and tracking. Simulation results show the effectiveness of the proposed method.

Most of the current vision sensor-based target detection is suitable for good weather conditions. Adverse weather conditions, especially foggy environments, significantly reduce visibility, which seriously affects the target detection performance. To improve driving safety in foggy environments, this paper proposes an improved YOLOX-based vehicle and pedestrian detection method in foggy environments. The method is based on the advanced YOLOX network model and introduces an attention mechanism in the feature extraction network to enhance the network's extraction of target features in foggy images. Some images in the training dataset are fogged to supplement the target-specific features in foggy environments and improve the robustness of the target detection network in foggy environments. The idea of migration learning is used in the training process to save training time and optimize the training effect. The experimental results show that the target detection method proposed in this paper has significantly improved the detection performance of vehicles and pedestrians in the foggy environment, with an 11.35% improvement in mAP, and the detection effect is better than the GCANet image defogging method. The effectiveness of the method improvement is proved.

Wireless sensor network has advantages of rapid deployment, self-organization, strong concealment, and high fault tolerance, etc. and it is very suitable for military reconnaissance. Considering the requirements of battlefield detection environment for men, gunmen and vehicles, a small-scale wireless sensor network(WSN) system is designed under laboratory conditions to simulate the target detection. According to the characteristics of different targets and zoning, the target feature extraction and detection methods are proposed for the signals of infrared sensors, magnetic sensors and sound sensors. A simple and efficient fixed threshold algorithm based on variance for infrared and magnetic resistance signals is put forward and a data fusion algorithm of D-S evidence theory is used to identify the targets. The binary detection algorithm of multi-sensor data fusion is presented for single target location and tracking. Finally, the experimental results have shown that the designed system can achieve reliable target detection, recognition, location and tracking in the laboratory.

The cucumber disease images obtained from natural environments often contain noise such as variations in lighting and soil conditions, which significantly impact the accuracy of disease recognition. Additionally, existing detection models require large memory footprints, making real-time cucumber disease detection challenging. To address the challenges associated with detecting small targets in cucumber disease images, this study presents an algorithm named CucumberDet, which integrates global information perception and feature fusion. Initially, we employ the Swin Transformer as the backbone network for RetinaNet to augment the primary network’s feature extraction capabilities, thus enhancing its ability to extract information globally. Subsequently, to strengthen the network’s detection capabilities, especially for remote and small targets, we introduce a highly effective Small Target Feature Fusion Module (SFFM) to meticulously integrate detailed data of small targets into shallow feature maps. Finally, to further refine the network’s capability to identify multi-scale targets and facilitate the flow of low-level feature information to high-level features, we introduce a novel Multi-level Feature Adaptive Fusion Module (MFAFM). Encouraging detection results are obtained across three distinct datasets, with experimental findings on a self-compiled cucumber disease image dataset revealing that our proposed algorithm improves detection accuracy by 6.8% compared to the original RetinaNet baseline network. The proposed model achieves an mAP of 92.5%, with a parameter count of 38.39 million and a frame per second (FPS) rate of 23.6, underscoring its superior performance in detecting small targets and demonstrating its effectiveness across various application scenarios.

The precession feature of space target provides an effective approach for target recognition. However, with the requirements that target detection and tracking to be completed successfully before the processing of precession features extraction can be implemented, the existing methods demand radar resources allocation for target detection, tracking and feature extraction, respectively, thus reducing the radar efficiency. In this paper, by establishing a feedback loop between precession feature extraction and TBD (track before detect) of target, a cognitive feature extracting method is proposed. And the sliding-type scatterer model is used for describing rotationally symmetric target. With the proposed method, the precession feature parameters of target can be extracted concurrent to implementing the target detecting and tracking. Simulation results show the effectiveness of the proposed method.

ConvNeXt-YOLOX model, an improved deep learning algorithm based on the YOLOX neural network, is proposed for surface mount technology (SMT) production line printed circuit board assembly (PCBA) solder joint defect detection with high accuracy and speed. The improved model employs the classification neural network ConvNeXt to enhance the feature extraction capability of the YOLOX backbone feature extraction network, thus enhancing the target detection accuracy and speed as well as the small target detection capability while maintaining the same number of parameters. A solder joint defect detection dataset containing 759 PCBA solder joint defect images is constructed, on which the improved model ConvNeXt-YOLOX, the original model YOLOX, and the lightweight model YOLOX-s are trained. Subsequently, the training results of the three models are compared. The comparison shows that the mean average precision (mAP) of the improved model ConvNeXt-YOLOX is 97.21%, which is 0.82% and 3.02% higher than that of YOLOX and YOLOX-s, respectively, while the mAP at (0.5:0.95) is increased from 76.3 to 77.5. Moreover, the detection speed is increased from the original 27.06 to 27.88 frames/s. In summary, the improved model ConvNeXt-YOLOX has strong small target feature extraction and detection capabilities, which are consistent with the actual requirements of solder joint defect detection.

In recent years, knowledge graphs (KGs) have been widely applied in various domains for different purposes. The semantic model of KGs can represent knowledge through a hierarchical structure based on classes of entities, their properties, and their relationships. The construction of large KGs can enable the integration of heterogeneous information sources and help Artificial Intelligence (AI) systems be more explainable and interpretable. This systematic review examines a selection of recent publications to understand how KGs are currently being used in eXplainable AI systems. To achieve this goal, we design a framework and divide the use of KGs into four categories: extracting features, extracting relationships, constructing KGs, and KG reasoning. We also identify where KGs are mostly used in eXplainable AI systems (pre-model, in-model, and post-model) according to the aforementioned categories. Based on our analysis, KGs have been mainly used in pre-model XAI for feature and relation extraction. They were also utilised for inference and reasoning in post-model XAI. We found several studies that leveraged KGs to explain the XAI models in the healthcare domain.

ProMvSD: Towards unsupervised knowledge graph anomaly detection via prior knowledge integration and multi-view semantic-driven estimation

As Data Flows Surge Data Infrastructure Groans

Recommender system (RS) devotes to predicting user preference to a given item and has been widely deployed in most web-scale applications. Recently, knowledge graph (KG) attracts much attention in RS due to its abundant connective information. Existing methods either explore independent meta-paths for user-item pairs over KG, or employ graph neural network (GNN) on whole KG to produce representations for users and items separately. Despite effectiveness, the former type of methods fails to fully capture structural information implied in KG, while the latter ignores the mutual effect between target user and item during the embedding propagation. In this work, we propose a new framework named Adaptive Target-Behavior Relational Graph network (ATBRG for short) to effectively capture structural relations of target user-item pairs over KG. Specifically, to associate the given target item with user behaviors over KG, we propose the graph connect and graph prune techniques to construct adaptive target-behavior relational graph. To fully distill structural information from the sub-graph connected by rich relations in an end-to-end fashion, we elaborate on the model design of ATBRG, equipped with relation-aware extractor layer and representation activation layer. We perform extensive experiments on both industrial and benchmark datasets. Empirical results show that ATBRG consistently and significantly outperforms state-of-the-art methods. Moreover, ATBRG has also achieved a performance improvement of 5.1% on CTR metric after successful deployment in one popular recommendation scenario of Taobao APP.

Due to the cigarette targets are not obvious, quality of monitoring picture is not clear and other factors lead to slow detection and feature extraction difficulties for the actual monitoring process. This paper proposes a target detection scheme of cigarettes with using the fusion of Yolov4 target detection algorithm and feature extraction of the human body region algorithm based on HOG. This paper uses the Yolov4 target detection algorithm as the main framework, intending to shorten the time of cigarette detection and ensure the real-time monitoring process. To solve the problem of difficult feature extraction for cigarette targets detection and effectively reduce the CPU usage, it makes a preliminary check of the human area for the presence of smoke while adding the hog based human area extraction algorithm to the algorithm before the cigarette targets detection. Experiments show that the method can effectively solve the problem of cigarette targets detection in public places. Compared with the original faster region convolution algorithm, the detecting time of a single image slightly increases but still meets the requirements of practical engineering applications. On the other hand, the rate of detecting incorrectly is reduced by about 2% and the detection accuracy is significantly improved.

Thanks to the increasing availability of drug-drug interactions (DDI) datasets and large biomedical knowledge graphs (KGs), accurate detection of adverse DDI using machine learning models becomes possible. However, it remains largely an open problem how to effectively utilize large and noisy biomedical KG for DDI detection. Due to its sheer size and amount of noise in KGs, it is often less beneficial to directly integrate KGs with other smaller but higher quality data (e.g. experimental data). Most of existing approaches ignore KGs altogether. Some tries to directly integrate KGs with other data via graph neural networks with limited success. Furthermore most previous works focus on binary DDI prediction whereas the multi-typed DDI pharmacological effect prediction is more meaningful but harder task. To fill the gaps, we propose a new method SumGNN: knowledge summarization graph neural network, which is enabled by a subgraph extraction module that can efficiently anchor on relevant subgraphs from a KG, a self-attention based subgraph summarization scheme to generate reasoning path within the subgraph, and a multi-channel knowledge and data integration module that utilizes massive external biomedical knowledge for significantly improved multi-typed DDI predictions. SumGNN outperforms the best baseline by up to 5.54%, and performance gain is particularly significant in low data relation types. In addition, SumGNN provides interpretable prediction via the generated reasoning paths for each prediction. The code is available in Supplementary Material. Supplementary data are available at Bioinformatics online.

Due to background clutter in synthetic aperture radar (SAR) images, the detection of dense ship targets suffers from a low detection rate, high false alarm rate, and high missed detection rate. To address this issue, an FSM-DFF-YOLOv5+Confluence algorithm is proposed in this paper for the detection of near-shore ship targets in SAR images with complex backgrounds. First, based on the YOLOv5 target detection algorithm, two improvements are made in the feature extraction network: feature refinement and multi-feature fusion; in the feature extraction network, deformable convolutional neural networks are adopted to change the position of the target sampling points of the convolution to improve the feature extraction capability of the target and the detection rate of ship targets in SAR images with a complex background; in the multi-feature fusion network structure, cascading and parallel pyramids are used in the multi-feature fusion network to realize feature fusion at different levels; the visual perceptual field of feature extraction is expanded by using null convolution to enhance the adaptability of the network to detect near-shore multi-scale ship targets with complex backgrounds and reduce the false alarm rate of ship target detection in SAR images with complex environments. In this way, the DFF-YOLOv5 near-shore ship target detection algorithm is established. Meanwhile, to address the problem of missed detection in near-shore dense ship target detection, this paper adds rectangular convolution kernels to the convolution of the feature extraction network to better realize the feature extraction of dense ship targets in SAR images with complex backgrounds. Besides, the Confluence algorithm instead of non-maximum suppression is used in the prediction stage. Through experiments on the constructed complex background near-shore ship detection dataset, it is indicated that the average accuracy of the FSM-DFF-YOLOv5+Confluence detection algorithm reaches 88.96%, and the recall rate reaches 88.80%.