Search

As a research hotspot of computer vision, crowd counting methods have achieved success in natural images. But crowd counting in aerial images are rarely explored, and existing methods do not perform well because of the higher resolution, smaller object scale and more complex scene. Therefore, this paper proposes a lightweight dual-task network (LDNet) for crowd counting, which only uses bifurcated structure to overcome these new challenges in aerial images without complicated pipelines. To realize this, a complete but efficient Guidance Branch is proposed to assist Counting Branch in fitting crowd distribution. Furthermore, a scene attention mechanism is used to consider the complex scene information, which are never considered by existing methods. Our LD-Net outperforms existing methods on aerial crowd counting dataset (Visdrone), and gets better or comparable results on natural crowd counting datasets (UCF_CC_50, UCF_QNRF, ShanghaiTech Part A).

Several algorithms have been developed to simplify buildings based on their local structure in past decades. However, different local structures are defined for certain purposes, and no algorithm can appropriately simplify all buildings. We propose a combined building simplification approach based on local structure classification and backtracking strategy. In this approach, local structures are classified and their based operations are defined by considering the buildings’ orthogonal and non-orthogonal features. Each building is simplified to target scale with a selected local-structure-based operation progressively scale-by-scale. Rules are built to support the selection of local-structure-based operations with a binary decision tree, and a backtracking strategy is used when an invalid operation is applied. When the building is too small or the evaluation shows that it cannot be simplified based on local structures, template matching or enlargement algorithms are applied to simplify the building. A dataset (1:10k) collected from the Ordnance Survey was used for the experiment and simplified scale of 1:25k. Results satisfied legibility constraints and the change in area, orientation and position of simplified buildings are controlled within certain range by comparing with the results generated based on other four simplification algorithms. Possible use of our approach for continuous scale transformation of buildings is also discussed.

The Vision Meets Drone Object Detection in Image Challenge (VisDrone-DET 2020) is the third annual object detector benchmarking activity. Compared with the previous VisDrone-DET 2018 and VisDrone-DET 2019 challenges, many submitted object detectors exceed the recent state-of-the-art detectors. Based on the selected 29 robust detection methods, we discuss the experimental results comprehensively, which shows the effectiveness of ensemble learning and data augmentation in drone captured object detection. The full challenge results are publicly available at the website http://aiskyeye.com/leaderboard/.

Crowd counting on the drone platform is an interesting topic in computer vision, which brings new challenges such as small object inference, background clutter and wide viewpoint. However, there are few algorithms focusing on crowd counting on the drone-captured data due to the lack of comprehensive datasets. To this end, we collect a large-scale dataset and organize the Vision Meets Drone Crowd Counting Challenge (VisDrone-CC2020) in conjunction with the 16th European Conference on Computer Vision (ECCV 2020) to promote the developments in the related fields. The collected dataset is formed by $3,360$ images, including $2,460$ images for training, and $900$ images for testing. Specifically, we manually annotate persons with points in each video frame. There are $14$ algorithms from $15$ institutes submitted to the VisDrone-CC2020 Challenge. We provide a detailed analysis of the evaluation results and conclude the challenge. More information can be found at the website: \url{http://www.aiskyeye.com/}., The method description of A7 Mutil-Scale Aware based SFANet (M-SFANet) is updated and missing references are added

Information on spatial distribution of buildings must be explored as part of the process of map generalization. A new approach is proposed in this article, which combines building classification an...

Spatial conflicts can remain unresolved during map generalization if only displacement is used, especially for urban building maps. To solve all possible spatial conflicts during urban building map generalization, we propose a collaborative displacement method, combining aggregation, elimination, and constrained reshape. After applying an improved vector field-based displacement for solution of initial conflicts, two types of sequential conflicts are detected and evaluated to identify additional displacement solutions based on a trial and error strategy. If no displacement solution is available, aggregation, elimination, or constrained reshape are selected and applied for the unsolved sequential conflicts based on cartographic rules. For a more reasonable generalization result, an improved constrained reshape approach is also introduced for buildings in conflicts along roads. Two data sets with scales of 1:5000 and 1:25 000 for urban building map were selected to validate this approach. The results indicate that it is reasonable and feasible to solve all defined spatial conflicts in urban building map generalization using our proposed method.

Research on tree-type irrigation pipe networks is an important component of agricultural water-saving projects. The optimal design of tree-type irrigation pipe networks is a key aspect regarding the profitability of irrigated agriculture. Meanwhile, swarm intelligence optimization algorithms have good computational ability and can be applied to solve many optimization problems in agricultural engineering. To identify the lowest investment cost for a pipe network, this study defined the concept of an upper water node to ensure the connectivity of tree-type irrigation pipe networks, and therefore, improve the pipe network planning model without using preliminary network connection diagrams. In addition, this study proposed an improved chicken swarm optimization algorithm (Improved Chicken Swarm Optimization using Adaptive Search and Variation, ICSO-ASV), which was applied to solve 12 test functions of different dimensions. The test results show that, compared to the traditional chicken swarm algorithm and other algorithms in the control group, the ICSO-ASV algorithm could effectively improve the global search capability. Finally, the ICSO-ASV algorithm was used to plan and design 15-node and 40-node pipe networks. The calculation results show that the average investment costs of the two pipe networks generated by the ICSO-ASV algorithm were 42.20% and 31.09% lower than those generated by the traditional chicken swarm algorithm, which further verified the feasibility of applying ICSO-ASV to design tree-type irrigation pipe networks. Thus, the design method proposed in this study can solve the optimal problems of tree-type irrigation pipe networks with varying topologies. The optimal solutions can be generated automatically using the ICSO-ASV algorithm if essential parameters of the pipe network planning model are provided.

The performance of RFID networks can be optimized by RFID reader scheduling. This paper proposed a novel approach using an improved bat algorithm (IBA), which can optimize RFID networks. The IBA algorithm includes 2 improved mechanisms. In the proposed approach, all RFID readers are scheduled to work in appropriate sequence, which can greatly reduce RFID collisions. Experiments on two different RFID networks have been carried out to evaluate the effectiveness. Simulation results show that the proposed approach using IBA has better optimization precision than control algorithms.

Since the cloud storage technology has distribution, isolation and sharing characteristics, key management has become more and more difficulty. As a consequence, the research of key management technology has become a hot topic in recent years. In order to solve the untrustworthiness of cloud storage server provider, the complexity, security of key and some other issues under cloud storage environment, this paper proposed a (n+1,s+1) key management technology on the basis of Shamir (n,s) and then this technology was used in cloud storage system. The main idea of the technology is that the whole key spited into n+1 parts and distributed to the different cloud storage service providers and data owner to manage the sub-keys. Through the performance analysis, in this technology, the data owner is the core of the management process. The technology is more secure than the traditional technology and solves the untrustworthiness of cloud storage server provider based on the premise of reducing the burden of data owner.