Dongil Han

Senti-lexicon and improved Naïve Bayes algorithms for sentiment analysis of restaurant reviews

The existing senti-lexicon does not sufficiently accommodate the sentiment word that is used in the restaurant review. Therefore, this thesis proposes a new senti-lexicon for the sentiment analysis of restaurant reviews. When classifying a review document as a positive sentiment and as a negative sentiment using the supervised learning algorithm, there is a tendency for the positive classification accuracy to appear up to approximately 10% higher than the negative classification accuracy. This creates a problem of decreasing the average accuracy when the accuracies of the two classes are expressed as an average value. In order to mitigate such problem, an improved Naive Bayes algorithm is proposed. The result of the experiment showed that when this algorithm was used and a unigrams+bigrams was used as the feature, the gap between the positive accuracy and the negative accuracy was narrowed to 3.6% compared to when the original Naive Bayes was used, and that the 28.5% gap was able to be narrowed compared to when SVM was used. Additionally, the use of this algorithm based on the senti-lexicon showed an accuracy that improved by a maximum of 10.2% in recall and a maximum of 26.2% in precision compared to when SVM was used, and by a maximum of 5.6% in recall and a maximum of 1.9% in precision compared to when Naive Bayes was used.

Image contrast enhancement based on the piecewise-linear approximation of CDF

In this paper, we present an image contrast enhancing method based on the piecewise linear approximation of the cumulative density function (CDF). Reduced hardware is required for CDF calculation, since the number of bins is greatly reduced thanks to this approximation. With the proposed method, various contrast enhancing effects can be implemented such as histogram equalization, partial or entire histogram modification.

A motion adaptive 3-D de-interlacing algorithm based on the brightness profile pattern difference

A novel motion adaptive de-interlacing algorithm is described. The accuracy of motion detection between fields is significantly improved by difference calculation with the brightness profile pattern instead of the simple pixel differences. The performance of a pure spatial interpolation is also enhanced by directional filtering combined with a hierarchical rule based detection scheme. The proposed algorithm is successfully implemented in a (H)DTV video display processor.

Real-time color gamut mapping method for digital TV display quality enhancement

A novel real-time color gamut mapping method is described. The color gamut mapping method that is used for enhancing the color reproduction quality between PC monitor ad printer devices is adopted for digital TV display quality enhancement. The high definition digital TV display devices operate at the clock speed of around 70 MH to 150 MHz and permit several nano seconds for real-time color gamut mapping. Thus, the concept of three-dimensional reduced resolution look-up table is introduced for real-time processing. The required hardware can be greatly reduced by look-up table resolution adjustment. The proposed hardware architecture is successfully implemented in FPGA and ASIC and also successfully adopted in digital TV display quality enhancement purposes.

Real-time fire detection using camera sequence image in tunnel environment

In this paper, we proposed image processing technique for automatic real time fire and smoke detection in tunnel environment. To avoid the large scale of damage of fire occurred in the tunnel, it is necessary to have a system to minimize and to discover the incident as fast as possible. However it is impossible to keep the human observation of Closed-Circuit Television (CCTV) in tunnel for 24 hour. So if the fire and smoke detection system through image processing can warn fire state, it will be very convenient, and it can be possible to minimize damage even when people is not in front of monitor. The fire and smoke detection is different from the forest fire detection as there are elements such as car and tunnel lights and others that are different from the forest environment so that an indigenous algorithm has to be developed. The two algorithms proposed in this paper, are able to detect the exact position, at the earlier stage of incident. In addition, by comparing properties of each algorithm throughout experiment, we have proved the validity and efficiency of proposed algorithm

A novel stereo matching method for wide disparity range detection

This paper describes a real-time stereo depth measurement image processing system. This system uses Xilinx Virtex-II Series XC2V3000 FPGA and generates 8-bit sub-pixel disparities on 640 by 480 resolution images at video rate (60 frames/sec) with maximum disparity ranges of up to 128 pixels. The implemented stereo matching algorithm finds a minimum of window-based sum of absolute difference (SAD) operation. And the preprocessing, scale transformation and final stage compensation technique are adopted for maximizing the wide disparity range detection. The proposed vision system is suitable for real-time range estimation and robot navigation applications.

A cost effective color gamut mapping architecture for digital TV color reproduction enhancement

A cost effective three-dimensional color gamut mapping architecture is described. The conventional three-dimensional reduced resolution look-up table is considered and the concept of three-dimensional reduced resolution difference look-up table is introduced for cost effective and real-time color gamut mapping. The overall architecture uses one-dimensional memory decomposition of three-dimensional gamut mapping look-up table and three-dimensional interpolation and simple adding operation for generating the final gamut mapped colors. The required computational cost is greatly reduced by look-up table resolution adjustment and further reduced by the gamut mapping rule modification. The proposed architecture greatly reduces the required memory size and hardware complexity compared to the conventional methods and it is suitable for real-time applications. The proposed hardware architecture is suitable for FPGA and ASIC implementation and could be applied to the real-time display quality enhancement purposes.

Development of an automatic adjustment system for integrated tube components

The design of a multiprocessor-based automatic adjustment system for ITCs (integrated tube components) used in color picture tube fabrication is described. A rule-based hierarchical control approach is proposed to resolve the problem caused by uncertainties and nonlinearities in ITC processes. The electronic control system consists of three subsystems, each of which uses 32 bit microprocessors. Various image processing algorithms were incorporated into the control system for fast and reliable operation by using online processing units and a ral-time frame grabber. The general properties and difficulties of the ITC process are described, and the system configuration is outlined. Vision sensing and control algorithms are considered.<<ETX>>

Design and VLSI implementation of high-performance face-detection engine for mobile applications

In this paper, we proposes a novel hardware architecture of face-detection engine for mobile applications. We used MCT(Modified Census Transform) and Adaboost learning technique as basic algorithms of face-detection engine. We have designed, implemented and verified the hardware architecture of face-detection engine for high-performance face detection and real-time processing. The face-detection chip is developed by verifying and implementing through FPGA and ASIC. The developed ASIC chip has advantage in real-time processing, low power consumption, high performance and low cost. So we expect this chip can be easily used in mobile applications.

Development of early tunnel fire detection algorithm using the image processing

To avoid the large scale of damage of fire occurred in the tunnel, it is necessary to have a system to minimize and to discover the incident fast. However it is impossible to keep the human observation of CCTV in tunnel for 24 hour. So if the fire and smoke detection system through image processing warn fire state, it can be very convenient, and it can be possible to minimize damage even when people is not in front of monitor. In this paper, we proposed algorithm using the image processing, which is an early detection of the fire and smoke occurrence in the tunnel. The fire and smoke detection is different from the forest fire detection as there are elements such as car and tunnel lights and others that are different from the forest environment so that an indigenous algorithm has to be developed. The two algorithms proposed in this paper, are able to detect the exact position, at the earlier stay of detection. In addition, by comparing properties of each algorithm throughout experiment, we have proved the propriety of algorithm.