Suree Pumrin

Video processing and analysis for surveillance applications

Presently researches in networked surveillance system grow continuously and substantially. One reason is because of the insecurity incidents such as terrorism acts in Thailand and many countries around the world. This results in the need of intelligent surveillance and monitoring system consisting of real-time image capture, transmission, processing, and surveillance information understanding. This information will be vital to people safety and indeed national security. Video Information Processing and Analysis for Surveillance System is a pilot project to realize such intelligent surveillance system developed by Thai researchers. The developed system consists of 6 parts, namely, super-resolution image reconstruction, suspect detection, suspect tracking, suspect appearance extraction, suspect activity analysis and level of awareness decision. The output information from the proposed system is suspect features, suspect activity pattern and direction, and level of suspicious incident. This would be very useful for the users to further analyze this information to use in practical surveillance system.

Elastic Net for solving sparse representation of face image super-resolution

Super-resolution is very important in recognizing suspects face in video surveillance system. In this paper, we present an improvement of image super-resolution based on sparse signal representation. The issue of how to deal efficiently with sparse feature has great significance on the quality improvement of generated high resolution image. We propose to use Elastic net to solve sparse representation of image super-resolution process. Elastic Net will compromise between Lasso and Ridge regression to find the best correlated patch between low-resolution and high-resolution image. Experiments demonstrate that using Elastic Net reconstructed high-resolution images have better color and texture quality. It also gives smaller value of Root Mean Square Error (RMSE) than Lasso and other conventional methods. Small RMSE means more accurate when recognizing face.

In-vehicle carbon dioxide concentration in commuting cars in Bangkok, Thailand

ABSTRACT It is known that in-vehicle carbon dioxide (CO2) concentration tends to increase due to occupant exhalation when the HVAC (heating, ventilation, and air conditioning) air is in recirculation mode. Field experiments were conducted to measure CO2 concentration during typical commute in Bangkok, Thailand. The measured concentrations agreed with the concentration predicted using first-order mass balance equation, in both recirculating and outside air modes. The long-term transient decay of the concentration when the vehicle was parked and the HVAC system was turned off was also studied. This decay was found to follow Fickian diffusion process. The paper also provides useful operational details of the automotive HVAC system and fresh air ventilation exchange between cabin interior and exterior. Implications: Drivers in tropical Asian countries typically use HVAC recirculation mode in their automobiles. This behavior leads to excessive buildup of cabin CO2 concentration levels. The paper describes the CO2 buildup in a typical commute in Bangkok, Thailand. Auto manufacturers can potentially take measures to alleviate such high concentration levels. The paper also discusses the diffusion of CO2 through the vehicle envelope, an area that has never been investigated before.

Elderly care notification system using hand posture recognition

This paper proposes a hand posture recognition as an assistive tool for elderly care. The system uses Haar-like feature, Principle Component Analysis (PCA), and simulation web server for detection, recognition, and alert by order. Our system interprets the hand sign captured by web camera and alerts a caregiver via smart phone. Besides, a caregiver can also periodically check needs of an elder from a smart phone. So a caregiver can do other tasks more efficiently because he/she need not to be with an elder at all times. The empirical results show satisfactory performance of PCA and contour detection with hand posture recognition.

The vision-based hand gesture recognition using blob analysis

As technology becomes the part of human life for decades, the relationship between human and computer called human-computer interaction (HCI) is important to study for improving the system to serve the human need. HCI can be applied in various areas including medical system which is valuable for the elder who is not able to walk or express the feelings by words. The intuitive approach is the development of algorithm by using hand gestures. This paper proposes a dynamic hand gesture recognition algorithm for elder people. The algorithm implements in a vision-based hand gesture recognition using optical flow and blob analysis to track six dynamic hand gestures and classify their meanings. The experiment provided good results for all six hand gestures in detection, tracking, and classification procedures.

1-D Integer Transform for HEVC Encoder Using DSP Slices on FPGA

This paper presents an FPGA architecture for the 1-D integer transform of the latest video coding standard, the High Efficiency Video Coding (HEVC). The design employs hard multipliers in dedicated DSP slices, which are already embedded into an FPGA die, to gain high throughput and save general purpose LUTs. The proposed architecture can support 4x4, 8x8, 16x16, and 32x32 transform. A multiplier sharing scheme is introduced to reduce the total number of required DSP slices in order to be able to fit the design onto a Spartan-3A FPGA. The design can reach a maximum throughput of 1,692 Msamples/s irrespective of the transform size, which is enough to encode 8K (7680x4320) videos at 30 fps. This work is a pioneer research that utilizes the dedicated multipliers on FPGAs in the design of the HEVC transform.

The Algorithm of Static Hand Gesture Recognition using Rule-Based Classification

Technology becomes a part of human lives for decades, especially in human—computer interaction (HCI) that considered as the important research area involving with an assistive technology and a medical system. Hand gesture is classified as an intuitive method for human to interact with the computer. It is useful for elderly people who cannot express their feelings by words. This paper proposed the hand gesture recognition technique for the elderly by using contour detection with convex hull feature extraction and rule-based classification. Vision-based hand gesture recognition is considered in classifying six hand gestures as lingual description. From the experimental results, the hand gesture system provides good detection and classification results for all the six static hand gestures.

Wireless indoor lamp status monitor

A device capable of data logging on/off status of all lamps in a house or office room for the purpose of monitoring energy consumption is described. It is based on 3 components which are a CMOS digital camera, a card type computer board and a WiFi module. An image of the room is taken once every minute and transformed to a two level black and white version from which a simple lamp detection is operated. Since the camera location is fixed, an initialization step can be implemented to reduce the candidate regions and avoid confusion due to reflection or window light. Together with the known lamp type, the energy consumption of each lamp as well as their total sum can be plot almost near real time through internet. Compared to other techniques, this approach is very simple and does not require in line installation and the estimation error can be within 10%.