A5087558241- Advanced Vision and Imaging
- Advanced Image and Video Retrieval Techniques
- Video Analysis and Summarization
- Image Retrieval and Classification Techniques
- Video Surveillance and Tracking Methods
- Image and Object Detection Techniques
- Face and Expression Recognition
- Image Enhancement Techniques
- Advanced Measurement and Detection Methods
- Industrial Vision Systems and Defect Detection
- Robotics and Sensor-Based Localization
- Anomaly Detection Techniques and Applications
- Optical measurement and interference techniques
- Face recognition and analysis
- Multimedia Communication and Technology
- Autonomous Vehicle Technology and Safety
- Music and Audio Processing
- Gaze Tracking and Assistive Technology
- Biometric Identification and Security
- Vehicle License Plate Recognition
- Innovation in Digital Healthcare Systems
- Color Science and Applications
- Advanced Data Compression Techniques
- Chaos-based Image/Signal Encryption
- Advanced Image Processing Techniques
Hanyang University
2013-2022
Anyang University
2015-2019
LG (United States)
2012
The University of Texas at Dallas
1994-1995
University of Dallas
1991
Hydrochromic materials have been actively investigated in the context of humidity sensing and measuring water contents organic solvents. Here we report a sensor system that undergoes brilliant blue-to-red colour transition as well 'Turn-On' fluorescence upon exposure to water. Introduction hygroscopic element into supramolecularly assembled polydiacetylene results hydrochromic conjugated polymer is rapidly responsive (<20 μs), spin-coatable inkjet-compatible. Importantly, found be suitable...
Urine tests are performed by using an off-the-shelf reference sheet to compare the color of test strips. However, tabular representation is difficult use and more prone visual errors, especially when color-swatches be compared spatially apart. Thus, making it distinguish between subtle differences shades on reagent pads. This manuscript represents a new arrangement arrays for urine strips (urinalysis). Reference swatches grouped in doughnut chart, surrounding each pad strip. The can...
Recent improvements in imaging sensors and computing units have led to the development of a range image-based human-machine interfaces (HMIs). An important approach this direction is use dynamic hand gestures for gesture-based interface, some methods been developed provide real-time skeleton generation from depth images gesture recognition. Towards end, we propose skeleton-based recognition method that divides geometric features into multiple parts uses gated recurrent unit-recurrent neural...
The separation of circulating tumor cells (CTCs) from the peripheral blood is an important issue that has been highlighted because their high clinical potential. However, techniques depend solely on tumor-specific surface molecules or just larger size CTCs are limited by heterogeneity. Here, we present a slanted weir microfluidic device utilizes and deformability to separate them unprocessed whole blood. By testing its ability using highly invasive breast cancer cell line, our achieved 97%...
Camera-based blind-spot detection systems improve the shortcomings of radar-based for accurately detecting position a vehicle. However, as with many camera-based applications, performance is insufficient in low-illumination environment such at night. This problem can be solved augmented nighttime images training data but acquiring them and annotating additional are cumbersome tasks. Therefore, we propose framework that converts daytime into synthetic using generative adversarial network...
An automated vision system that inspects brake shoes for rolling stock is proposed. The consists of two modules, namely, one image acquisition and another analysis. first module placed under the railway tracks automatically captures images using digital cameras as trains pass module. captured train information are then transferred into a database. Three specifications, thickness, any unbalanced wear on shoes, distances between wheels measured by second analysis Shadow regions defined...
The conventional method of estimating the rotation angle a pattern using principal axes is not suitable for circular symmetric patterns since their eigenvalues are similar in both directions. In paper, robust phase information Zernike moments presented. experimental results show that proposed estimates more accurately than method, even presence noise.
Eye detection is a critical task in many applications, especially for portable single user 3D display systems of the non-glasses type. However, existing methods either have high computational complexity or low rate. In addition, these do not work with those using eye glasses devices. this paper, we propose rapid method We novel operator that measures contrast features to compute difference intensity between iris region and its neighboring regions. The probability an identified area being...
Eye gaze tracking system has been widely researched for the replacement of conventional computer interfaces such as mouse and keyboard. In this paper, we propose long range binocular eye that works from 1.5 m to 2.5 with allowing a head displacement in depth. The 3D position user's is obtained two wide angle cameras. A high resolution image captured using pan, tilt, focus controlled narrow camera. angles maneuvering pan tilt motor are calculated by proposed calibration method based on...
We propose a method of detecting foreign objects in packaged foods with irregular texture patterns using one-class classification method. For reliable detection X-ray images, the contrast image is enhanced by reducing intensity food substrate. Since type and size cannot be known advance, we employ to discriminate from images. Foreign diverse types sizes were implanted some dry such as instant ramen, macaroni, spaghetti. real-time processing max-min difference mask operation utilized for...
A lane detection system using around view monitoring (AVM) images is presented in this paper. To provide safe driving condition, many approaches have been proposed. However, previous cannot detect lanes stably low visibility condition such as foggy or rainy days because of the use frontal camera. The proposed uses ego-vehicle's surrounding road information to overcome problem. method can be split into two stages: generation AVM from four fisheye cameras and images. generate images, we...
An ever increasing number of registered trademarks has created greater demand for an automatic trademark retrieval system. We present a method such system based on the image content using shape features. Zernike or pseudo-Zernike moments are employed as feature set. To retrieve similar shapes, we take into account visually salient features that dominantly affect global and ignore their minor detail. Experimental results database 3,000 images demonstrate proposed retrieves which agree well...
In the vision-based remote gaze tracking systems, most challenging topics are to allow natural movement of a user and increase working volume distance system. Several eye estimation methods considering have been proposed. However, their narrow close. this paper, we propose novel 2-D mapping-based method that allows large-movement user. Conventional utilize mapping function between calibration points on screen pupil center corneal reflection (PCCR) vectors obtained in step. PCCR associated...
We present a practical pattern recognition system that is invariant with respect to translation, scale and rotation of objects. The also insensitive large variations the threshold used. As feature vectors, Zernike moments are used we compare them Hu's seven moment invariants. For machine vision system, three key issues discussed: normalization, fast computation moments, classification using k-NN rule. testing results, recognizes set 62 alphanumeric machine-printed characters different sizes,...
We introduce an automatic video summarizing tool (AVST) for a personal recorder. The utilizes MPEG-7 visual descriptors to generate index summary. resulting generates not only preview of movie but also allows non-linear access with thumbnails. In addition, the supports searching shots similar desired one within saved sequences. Moreover, simple shot-based editing can readily be achieved using generated index.
Angular radial transform (ART), which is the region-based shape descriptor of MPEG-7, has desirable properties for representing information in a small number features with no redundancy. However, order ART to be useful, especially limited computing environments, computational cost must greatly reduced. In this paper, we derive symmetric/antisymmetric from basis functions and present fast efficient method compute coefficients using these properties. The proposed significantly reduces...
Counting the amount of pedestrian flow is an important task for video surveillance applications. Most previous methods counting employed model-based detection or top-view cameras to measure flow. However, those approaches are difficult apply realistic applications because high complexity specific camera set up. In this paper, a pixel count based method proposed estimation. method, image features such as foreground pixels and motion vectors utilized clues find number people going through...
Video synopsis allows us to analyze security videos efficiently by condensing or shortening a long video into short one. To generate condensed video, moving objects (a.k.a. object tubes) in the are rearranged temporal domain using predefined objective function. The function consists of several energy terms which play important roles making visually appealing video. One terms, collision energy, creates bottleneck computation because it requires two tubes calculate degree between them....