- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2018 IEEE International Conference on Consumer Electronics - Asia (ICCE-Asia) - Jeju (2018.6.24-2018.6.26)] 2018 IEEE International Conference on Consumer Electronics - Asia (ICCE-Asia) - Image Matting Using Feature Combination and Guided Filter
摘要： This paper proposes an accurate image matting method based on k-nn (k-nearest neighbors) clustering using alpha matte. First, we extract features consisting of eight channels using color, position, texture, and information for each pixel. Second, clustering similar pixels with k-nn clustering using extracted features. Finally, the estimated alpha matte is refined by post-processing. Through the proposed method, accurate image matting is possible because it shows a mean square error (MSE) value of 0.4540, which is lower than the previous study.
关键词： Image matting,Defense Surveillance Technology,Alpha matting
更新于2025-09-23 15:23:52
-
Performance of whole-body 18F-FDG PET/CT as a posttreatment surveillance tool for sinonasal malignancies
摘要： Purpose To determine the diagnostic utility of posttreatment surveillance whole-body 18F-FDG PET/CT in detecting local tumor recurrence (R), regional lymph-node metastasis (LM), and distant metastasis (DM) in asymptomatic sinonasal cancer patients. Methods Eighty consecutive patients (53 men, 27 women; mean age, 60 years; range, 28–92 years) who had undergone 197 posttreatment whole-body 18F-FDG PET/CT examinations for sinonasal malignancies between January 2009 and August 2017 were retrospectively reviewed. 18F-FDG PET/CT findings were categorized as positive or negative for R, LM, and DM, separately. Outcomes of 18F-FDG PET/CT scans were compared with the final diagnosis confirmed by histological analysis or follow-up period for a minimum 12 months. The diagnostic accuracy of scans was calculated for each site using contingency tables. Impact on the management of 18F-FDG PET/CT examinations was additionally evaluated. Results 18F-FDG PET/CT scans identified 37/44 of local recurrences, 21/23 of LMs, and 30/37 of DMs. For local recurrence, sensitivity, specificity, positive predictive value, and negative predictive value were 84% (68–97%), 95% (80–100%), 84% (68–97%), and 95% (80–100%), respectively. For LM, the respective values were 91% (75–100%), 99% (83–100%), 91% (75–100%), and 99% (83–100%). For DM, the values were 81% (64–97%), 99% (85–100%), 97% (81–100%), and 96% (81–100%), respectively. 18F-FDG PET/CT accounted for a change in management of 85% patients with recurrences. Conclusions Whole-body 18F-FDG PET/CT is a suitable surveillance tool for sinonasal malignancies in detecting locoregional and distant recurrences in asymptomatic patients without any evidence of recurrence on regular follow-up and endoscopy during the posttreatment period.
关键词： Posttreatment,Recurrence,Surveillance,Sinonasal malignancies,18F-FDG PET/CT
更新于2025-09-23 15:23:52
-
[IEEE 2018 International Conference on Radar (RADAR) - Brisbane, Australia (2018.8.27-2018.8.31)] 2018 International Conference on Radar (RADAR) - Multiband Polarimetric SAR in Arctic Scenarios
摘要： In relation to a working group on future capabilities for applications in the Arctic region, the Danish Defence Acquisition and Logistics Organization (DALO) has conducted test campaigns with the multi-band, fully polarimetric F-SAR system owned by the German Aerospace Center (DLR) in order to explore the possibilities that advanced synthetic aperture radar (SAR) systems provide for surveillance, change detection, moving target identification and high resolution imaging. Examples of results and some preliminary conclusions are presented in this paper.
关键词： detection,SAR,polarimetry,surveillance,Arctic,recognition
更新于2025-09-23 15:23:52
-
[ACM Press the 4th ACM SIGSPATIAL International Workshop - Seattle, WA, USA (2018.11.06-2018.11.06)] Proceedings of the 4th ACM SIGSPATIAL International Workshop on Safety and Resilience - Safety and Resilience'18 - Energy-Optimal Path Planning for Solar-Powered UAVs Monitoring Stationary Target
摘要： In this paper, based on the Particle Swarm Optimization (PSO) Algorithms and a novel penalty function, a hybrid method is proposed to plan the optimal trajectories of solar-powered UAVs (SUAVs) for monitoring stationary target. The goal of the route planning is to obtain the maximum net energy when the SUAVs have accomplished a mission under various constraints, e.g., aircraft dynamic constraint and simultaneous arrival at the given destination. First, the target surveillance problem take into account of energy optimization is modeled detailed by formulating the energy harvesting, the energy consumption, the sensor coverage area, the space constraints, etc. Second, based on the mode of stationary target surveillance, the problem of path planning is converted to the nonlinear optimization problem with constraints, and then, by using penalty function method the constrained optimization problem will be transformed to an unconstrained optimization problem. Next, in consideration of the computational complexity of this problem, PSO with penalty function, a novel intelligent algorithm with the advantages of good stability and strong search ability is adopted to solve the optimization problem. Finally, the proposed method is demonstrated and compared with traditional method in the simulated scenario. The simulation results show that it is feasible for this proposed hybrid method to solve the problem of energy optimal path planning.
关键词： optimal trajectories,stationary target surveillance,penalty function,PSO,SUAVs
更新于2025-09-23 15:23:52
-
[IEEE 2018 IEEE International Conference on Imaging Systems and Techniques (IST) - Krakow (2018.10.16-2018.10.18)] 2018 IEEE International Conference on Imaging Systems and Techniques (IST) - Robust Estimation of Product Amount on Store Shelves from a Surveillance Camera for Improving On-Shelf Availability
摘要： This paper proposes a method to robustly estimate product amount on store shelves from a surveillance camera for improving on-shelf availability. We focus on changes of products on the shelves such as “product taken (decrease)” and “product replenished/returned (increase)”, and compute product amount by accurately accumulating them. The proposed method first detects change regions of products on the shelves in an image using background subtraction followed by moving object removal. The detected change regions are then classified into several classes representing the actual changes on the shelves such as “product taken” by using convolutional neural networks. Finally, the changes of products on the shelves are accumulated using classification results, and product amount on the shelves visible in the image is computed as on-shelf availability. Experimental results using two videos captured in a real store show that our method achieves success rate of 89.6% for on-shelf availability when an error margin is within one product. With high accuracy, store clerks can keep high on-shelf availability, enabling the improvement of business profit in retail stores.
关键词： product amount,image processing,on-shelf availability,convolutional neural network,surveillance camera,retail,background subtraction
更新于2025-09-23 15:22:29
-
Single Image Defogging Based on Illumination Decomposition for Visual Maritime Surveillance
摘要： Single image fog removal is important for surveillance applications and many defogging methods have been proposed, recently. Due to the adverse atmospheric conditions, the scattering properties of foggy images depend on not only the depth information of scene, but also the atmospheric aerosol model, which has more prominent influence on illumination in a fog scene than that in a haze scene. However, recent defogging methods confuse haze and fog, and they fail to consider fully about the scattering properties. Thus, these methods are not sufficient to remove fog effects, especially for images in maritime surveillance. Therefore, this paper proposes a single image defogging method for visual maritime surveillance. Firstly, a comprehensive scattering model is proposed to formulate a fog image in the glow-shaped environmental illumination. Then, an illumination decomposition algorithm is proposed to eliminate the glow effect on the airlight radiance and recover a fog layer, in which objects at the infinite distance have uniform luminance. Secondly, a transmission-map estimation based on the non-local haze-lines prior is utilized to constrain the transmission map into a reasonable range for the input fog image. Finally, the proposed illumination compensation algorithm enables the defogging image to preserve the natural illumination information of the input image. In addition, a fog image dataset is established for visual maritime surveillance. The experimental results based on the established dataset demonstrate that the proposed method can outperform the state-of-the-art methods in terms of both the subjective and objective evaluation criteria. Moreover, the proposed method can effectively remove fog and maintain naturalness for fog images.
关键词： Illumination Decomposition,Haze-Lines Prior,Image Defogging,Atmospheric Aerosol Model,Visual Maritime Surveillance
更新于2025-09-23 15:22:29
-
Multiview Layer Fusion Model for Action Recognition Using RGBD Images
摘要： Vision-based action recognition encounters different challenges in practice, including recognition of the subject from any viewpoint, processing of data in real time, and offering privacy in a real-world setting. Even recognizing profile-based human actions, a subset of vision-based action recognition, is a considerable challenge in computer vision which forms the basis for an understanding of complex actions, activities, and behaviors, especially in healthcare applications and video surveillance systems. Accordingly, we introduce a novel method to construct a layer feature model for a profile-based solution that allows the fusion of features for multiview depth images. This model enables recognition from several viewpoints with low complexity at a real-time running speed of 63 fps for four profile-based actions: standing/walking, sitting, stooping, and lying. The experiment using the Northwestern-UCLA 3D dataset resulted in an average precision of 86.40%. With the i3DPost dataset, the experiment achieved an average precision of 93.00%. With the PSU multiview profile-based action dataset, a new dataset for multiple viewpoints which provides profile-based action RGBD images built by our group, we achieved an average precision of 99.31%.
关键词： privacy-preserving surveillance,layer fusion model,real-time processing,Multiview action recognition,RGBD images,depth-based features
更新于2025-09-23 15:22:29
-
The Placement of Digitized Objects in a Point Cloud as a Photogrammetric Technique
摘要： The frequency of video-capturing collision events from surveillance systems are increasing in reconstruction analyses. The video that has been provided to the investigator may not always include a clear perspective of the relevant area of interest. For example, surveillance video of an incident may have captured a pre- or post-incident perspective that, while failing to capture the precise moment when the pedestrian was struck by a vehicle, still contains valuable information that can be used to assist in reconstructing the incident. When surveillance video is received, a quick and efficient technique to place the subject object or objects into a three-dimensional environment with a known rate of error would add value to the investigation. In addition, once the objects have been placed into the three-dimensional environment, the investigator would then be able to observe the physical evidence and environment from any perspective, including viewing and measuring what cannot be seen in the video perspective. In this research, the proposed photogrammetric technique of visually placing objects within three-dimensional laser scans will be evaluated. This research aims to quantify the rate of error of taking measurements of these objects to known fixed reference points both in and out of view of the camera, and provide an efficient technique that can be employed by reconstructionists using only one software package. As a result of this research, the authors have developed an expedient, less time-intensive photogrammetric technique for the placement of three-dimensionally scanned objects and environments. This technique can take less than half of the time of a conventional photogrammetric solution.
关键词： Photogrammetry,Video footage,3D scanning,Surveillance video,Laser scanning,Scanagrammetry,Accident reconstruction,Drivecam,Dashcam
更新于2025-09-23 15:22:29
-
[IEEE 2018 2nd International Conference on Trends in Electronics and Informatics (ICOEI) - Tirunelveli (2018.5.11-2018.5.12)] 2018 2nd International Conference on Trends in Electronics and Informatics (ICOEI) - SmartMobiCam: Towards a New Paradigm for Leveraging Smartphone Cameras and IaaS Cloud for Smart City Video Surveillance
摘要： Cloud and mobile computing domains have become quite popular these days. Besides, the ubiquitous smartphones are increasingly becoming a dominant platform for collaborative sensing, so we built an android application SmartMobiCam, which aids in forming the robust and scalable Smart city video surveillance system which is critical for upgrading the urban areas. The application obtains video feed, images and sound clips from the users and then uses cloud services for video enhancement and restoration of the content and provides it to all subscribers including police investigators. The application is explained with detailed architectural and technological choices. Performance analysis shows that the proposed application performs well and outperforms the traditional surveillance system.
关键词： waiting time,video enhancement,Smart city Surveillance,SmartMobiCam,Cloud Mobile Computing
更新于2025-09-23 15:22:29
-
Cognitive Cameras - [Opinion]
摘要： Today, computer vision can track cars, faces, and production processes as accurately as most people can. When there’s a lot of data to sift through, computer-vision models are better than people. But there are limits. Computers still need more time than a human to recognize a person or action. They can’t follow a person or object between multiple video cameras. They can be fooled easily. They can’t assign meaning to what they see. These are the limits engineers must overcome to make cameras more useful in manufacturing and in smart cities.
关键词： surveillance,image processing,adversarial attacks,computer vision,cognitive cameras
更新于2025-09-23 15:21:21