Consumer video surveillance systems often suffer from bad weather conditions, observed objects lose visibility and contrast due to the presence of atmospheric haze, fog, smoke. In this paper, we present an image defogging algorithm with color correction in HSV space for processing. We first generate a modified transmission map segmentation using multiphase level set formulation intensity (V) values. also estimate light The proposed method can significantly enhance foggy frames estimated map....

Ship collision avoidance involves helping ships find routes that will best enable them to avoid a collision. When more than two encounter each other, the procedure becomes complex since slight change in course by one ship might affect future decisions of other ships. Two distributed algorithms have been developed response this problem: Distributed Local Search Algorithm (DLSA) and Tabu (DTSA). Their common drawback is it takes relatively large number messages for coordinate their actions....

As vital transportation carriers in trade, ships have the advantage of stability, economy, and bulk capacity over airplanes, trucks, trains. Even so, their loss cost due to collisions other accidents exceed those any mode transportation. To prevent ship many ways been suggested, e.g., 1972 COLREGs which is regulation for preventing collision between ships. Technologically speaking, related studies conducted. The term “Ship domain” involves that area surrounding a navigator wants keep clear...

More than 90% of world trade is transported by sea. The size and speed ships rapidly increasing in order to boost economic efficiency. If collide, the damage cost can be astronomical. It very difficult for officers ascertain routes that will avoid collisions, especially when multiple travel same waters. There are several ways prevent ship such as lookouts, radar, VHF radio. advanced methodologies, domain, fuzzy theory, genetic algorithm, have been proposed. These methods work well one-on-one...

We present a fully digital auto-focusing (FDAF) system with automatic focusing region selection and priori estimated dataset of circularly symmetric point-spread functions (PSFs). The proposed approach provides realistic, unsupervised PSF estimation by analyzing the entropy edge information in automatically selected region. main advantage is fast robust defocusing due to simply selecting optimal small, homogeneous region-ofinterest. FDAF consists functional units; i) selection, ii)...

In this paper, a novel image enhancement system for wide-angle lens camera is presented. The proposed consists of; i) distortion correction using space-varying interpolation kernels and ii) restoration based on the local self-similarity. process geometric produced by results in radial artifacts caused non-linear resampling. To reduce such artifacts, algorithm uses derived from calibration data. corrected further enhanced self-example-based restoration. Experimental demonstrate method can...

This study proposes an outlier detection system based on the visual-physiology multimodal data for a Korean reality TV show, “Perfect Paper.” The goal of this is to provide program production team and master ceremonies with real-time facial expressions bio-signal analysis results participants. Using information, can easily identify participants unusual behavior. We propose insole-type hardware that measures wearer’s skin conductivity, temperature, motion. also suggest dynamic time warp-based...

This paper presents an efficient method for securing navigation performance by suppressing divergence risk of LiDAR SLAM through a newly proposed geometric observability analysis in three-dimensional point cloud map. For this, characteristics are introduced that quantitatively evaluate the quality distribution features. To be specific, this study adapts 3D matrix and associated condition number developing numerical benefit. In extensive application, we implemented path planning which...

This paper explores the research topic of enhancing reliability unmanned mobile exploration using LiDAR SLAM. Specifically, it proposes a technique to analyze waypoints where 3D SLAM can be smoothly performed in potential areas and points there is risk divergence navigation estimation. The goal improve performance by presenting method that secures these candidate regions. analysis employs geometric observability matrix its condition number discriminate waypoints. Subsequently, discriminated...

Since the developments of a navigation system that relies mostly on real flight experiments in GNSS-challenged, urban environments are expensive, verification and modification process prototype through implementing its simulator takes great advantages. In this study, objective under consideration assumes range/INS integration with an GIS map data GNSS-denied condition. Most fundamentally, developing distributed components, Unity3D's game engine is adapted as useful tool for fast processing...

In this paper, we introduce a pose estimation system which is called MPS, where magnetic field employed as an information delivering medium. It consists of transmitter part and receiver part. The circular coil with driving resonator circuit that generates AC signal pre-designed frequency. We use multiple transmitters have different carrier frequencies. Each covers the navigation area overlaps in space. A 3-axis magnetometer measures summed flux density, signals are separated FDMA. Magnetic...

최근접점과 최근접시간에 기초를 둔 종래 선박충돌회피 알고리즘은 국제해상충돌예방규칙(COLREG, International Regulations for Preventing Collisions at Sea, 1972)에 의한 선박간의 항법규정을 고려하지 않는다는 문제점이 있다. 본 논문에서 제안하는 조우선박의 상대방위를 측정함으로써 자선이 피항선인가 유지선이 될 것인가를 판단한다. 피항시점을 결정하기 위하여 타원형의 선박안전영역 모델을 이용하여 타선과 위험한 위치에 도달하는 지점과 시간을 계산함으로써 피항을 수행한다. 두 선박의 위치, 속도 및 방위를 임의로 생성하여 타원모델에 충돌회피 과정을 시뮬레이션 한다. 또한 실제적인 해양사고 발생 자료를 충돌위험 분석 선박간 충돌회피를 위한 방안을 제시한다. 제안한 안개 기상악화의 경우 시계가 제한되거나 야간항해 중 항해등의 식별이 어려워 피항선과 유지선의 판단이 불확실한 경우에도 국제해상충돌예방규칙에 동작에 효과적으로 적용할 수 On...

An extended fisheye lens model is presented to control the size ratio between distorted and virtually undistorted images based on orthographic projection. The optimum derived correct barrel distortion of a so that maximum amount peripheral region reconstructed with minimum visual distortion. geometric correction generates an aliasing artifact in central jagging region. Based proposed model, novel image enhancement algorithm also remove artifacts geometrically corrected image. Experimental...

Dense prediction tasks are a fundamental class of problems in computer vision. As supervised methods suffer from high pixel-wise labeling cost, few-shot learning solution that can learn any dense task few labeled images is desired. Yet, current target restricted set such as semantic segmentation, presumably due to challenges designing general and unified model able flexibly efficiently adapt arbitrary unseen semantics. We propose Visual Token Matching (VTM), universal learner for tasks. It...

Convolutional neural networks (CNNs) have become a primary approach in the field of artificial intelligence (AI), with wide range applications. The two computational phases for every network are; training phase and testing phase. Usually, is performed on high-processing hardware engines, however, part still challenge low-power devices. There are several accelerators; such as graphics processing units field-programmable-gate-arrays (FPGAs). From design perspective, an efficient engine at...