Tensor Voting Based Text Localization in Natural Scene Images

Toan Dinh Nguyen,Jonghyun Park,Gueesang Lee IEEE 2010 IEEE signal processing letters Vol.17 No.7

<P>A new and efficient text localization method by tensor voting is proposed. Tensor voting is used to extract the text line information based on the observation that the text characters are situated close together and arranged in a line or on a smooth curve. The text line information is useful to reduce the false positive rate in region-based text localization methods. The experimental results obtained for different types of natural text images show that the proposed method successfully detects the text regions with a low false-positive rate.</P>

A Novel Interaction Method for Mobile Devices Using Low Complexity Global Motion Estimation

Toan Dinh Nguyen,김정환,양형정,이귀상,장준영,엄낙웅,김수형 한국전자통신연구원 2012 ETRI Journal Vol.34 No.5

A novel interaction method for mobile phones using their built-in cameras is presented. By estimating the path connecting the center points of frames captured by the camera phone, objects of interest can be easily extracted and recognized. To estimate the movement of the mobile phone, corners and corresponding Speeded-Up Robust Features descriptors are used to calculate the spatial transformation parameters between the previous and current frames. These parameters are then used to recalculate the locations of the center points in the previous frame into the current frame. The experiment results obtained from real image sequences show that the proposed system is efficient, flexible, and able to provide accurate and stable results.

Using Tensor Voting for Detecting Noise Regions in Text Images

Toan Nguyen Dinh,Jonghyun Park,GueeSang Lee 한국멀티미디어학회 2009 한국멀티미디어학회 학술발표논문집 Vol.2009 No.1

Many kinds of noises may present in natural text images such as streaks, shadows, cracks and even small objects which produce occluded text. To achieve good binarized text results, we need to detect and remove noise regions. The 3D tensor voting framework generates the surface saliency map in which the pixel saliency value is directly proportional to the region area in which this pixel belongs to. Since the noise regions are usually smaller than background and text regions, we use 3D tensor voting to detect them. The experimental results show that we can successfully detect several kinds of noises in natural text images.

Formability improvement and blank shape definition for deep drawing of cylindrical cup with complex curve profile from SPCC sheets using FEM

Nguyen, Duc-Toan,Dinh, Duy-Khoe,Nguyen, Hong-Minh Thi,Banh, Tien-Long,Kim, Young-Suk Springer-Verlag 2014 Journal of Central South University Vol.21 No.1

Efficient MPEG-4 to H.264/AVC Transcoding with Spatial Downscaling

Toan Dinh Nguyen,이귀상,June-Young Chang,Han-Jin Cho 한국전자통신연구원 2007 ETRI Journal Vol.29 No.6

Efficient downscaling in a transcoder is important when the output should be converted to a lower resolution video. In this letter, we suggest an efficient algorithm for transcoding from MPEG-4 SP (with simple profile) to H.264/AVC with spatial downscaling. First, target image blocks are classified into monotonous, complex, and very complex regions for fast mode decision. Second, adaptive search ranges are applied to these image classes for fast motion estimation in an H.264/AVC encoder with predicted motion vectors. Simulation results show that our transcoder considerably reduces transcoding time while video quality is kept almost optimal.

The Impact of Psychological and Environmental Factors on Consumers’ Purchase Intention toward Organic Food: Evidence from Vietnam

Dinh Toan NGUYEN,Dinh Chien TRUONG 한국유통과학회 2021 The Journal of Asian Finance, Economics and Busine Vol.8 No.1

The study investigates some psychological and environmental factors affecting the intention to purchase organic foods of consumers in the inner-city of Hanoi. Impact factors applied for the study include three psychological factors (health concern, environmental concern, consumer awareness of organic foods) and seven environment factors (family’s opinion, friends and colleagues’ opinion, influence of celebrities, expert’s opinion, social status, mass media, state’s encouragement). We analyzed research data from 396 consumers to measure the impacting level of these factors. The convenient sampling method was used to collect the research sample. The measurement applied a 5-point Likert scale classifying from 1–completely disagree to 5–completely agree. Based on previous studies, the research model was recommended. We had estimated the reliability of the scales through Cronbach’s Alpha and composite reliability. The research data is analyzed by using Structural Equation Model method (SEM). The findings of the study suggest that psychological factors (health concern, environmental concern, consumer awareness of organic foods) had a significantly positive influence on consumer’s purchase attention toward organic food. The results also revealed that environmental factors (family’s opinion, friends and colleagues’ opinion, influence of celebrities, expert’s opinion, mass media) were positively linked to consumer’s purchase attention toward organic food.

Text Line Segmentation in Handwritten Document Images Using Tensor Voting

NGUYEN, Toan Dinh,LEE, Gueesang The Institute of Electronics, Information and Comm 2011 IEICE transactions on fundamentals of electronics, Vol.ea94 No.11

<P>A novel grouping approach to segment text lines from handwritten documents is presented. In this text line segmentation algorithm, for each text line, a text string that connects the center points of the characters in this text line is built. The text lines are then segmented using the resulting text strings. Since the characters of the same text line are situated close together and aligned on a smooth curve, 2D tensor voting is used to reduce the conflicts when building these text strings. First, the text lines are represented by separate connected components. The center points of these connected components are then encoded by second order tensors. Finally, a voting process is applied to extract the curve saliency values and normal vectors, which are used to remove outliers and build the text strings. The experimental results obtained from the test dataset of the ICDAR 2009 Handwriting Segmentation Contest show that the proposed method generates high detection rate and recognition accuracy.</P>

Flexible MPEG-4 to H.264 Transcoder for Video Transmission with Effective Motion Information Reusing

Toan Nguyen Dinh,JaeMyung Yoo,Sungchan Park,GueeSang Lee,Young Chang,Han-Jin Cho 대한전자공학회 2007 ITC-CSCC :International Technical Conference on Ci Vol.2007 No.7

In this paper, we present a flexible MPEG-4@SP (with Simple Profile) to H.264 transcoder for transmitting video over a system which has different kinds of networks, various types of devices and video formats. From decision modes and motion vectors in MPEG-4 decoder, a set of prediction motion vectors (PMV) and limited decision modes are considered in H.264 encoder to reduce the transcoding time. To maintain a reasonable video quality, we apply an adaptive integer search range based on video picture characteristics to find the optimal motion vectors. We also present a novel spatial downscaling transcoding method in the transcoder for reducing bit rate and spatial resolution to meet the constrained transmission bandwidths and terminal devices capabilities. The simulation results show that our transcoder considerably reduces transcoding time compare to other methods while video quality is kept almost optimal.

Static response and free vibration of functionally graded carbon nanotube-reinforced composite rectangular plates resting on Winkler–Pasternak elastic foundations

Duc, Nguyen Dinh,Lee, Jaehong,Nguyen-Thoi, T.,Thang, Pham Toan Elsevier 2017 AEROSPACE SCIENCE AND TECHNOLOGY Vol.68 No.-

<P><B>Abstract</B></P> <P>In the present article, static response and free vibration of functionally graded carbon nanotube reinforced composite (FG-CNTRC) rectangular plate resting on Winkler–Pasternak elastic foundations using an analytical approach are studied. The rectangular plates are reinforced by single-walled carbon nanotubes (SWCNTs) which are assumed to be graded through the thickness direction with four types of distributions. The mathematical model of the FG-CNTRC plate is developed based on the first-order shear deformation plate theory (FSDT) and Hamilton principle. By using Navier solution, the governing equations are solved to obtain the central deflection and the natural frequency parameters. Several examples are verified to have higher accuracy than those from the previous method in the literature. Also, the effects of different parameters on static response and natural frequency of FG-CNTRC plate are highlighted by solving numerous examples. Finally, these new results may serve as benchmarks for future investigations.</P>

Improving formability of tube bending for a copper material using finite element simulation

Duc-Toan Nguyen,Dinh-Thanh Nguyen,김영석 대한기계학회 2015 JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY Vol.29 No.10

Bending tubes are key products in many industries. The geometric parameters of the bending process are considered according to Taguchi’sorthogonal array and then coupled with finite element simulation to predict and improve the formability of the tube bendingprocess for copper JIS25A material. Three parameters, namely, mandrel diameter, distance between mandrel rings, and distance from thetip of the mandrel bar to the center of the base die, are selected to study their effects on the quality of the bending process. The varianceanalysis shows that the effect distribution of each parameter to bending quality is determined, and optimal conditions are adopted to conductexperiments.