Tien Dzung Nguyen

Efficient and reliable camera based multiple-choice test grading system

This paper proposes a new idea for grading multiple-choice test which is based on a camera with reliability and efficiency. The bounds of the answer sheet image captured by the camera is first allocated using Hough transform and then skew-corrected into the proper orientation, followed by the normalization to a given size. Next, the tick mark corresponding to the answer for each question can be recognized by allocation of the mask which wraps the answer area. The experimental results showed that the proposed system has achieved significant improvement in performance in terms of accuracy, reliability, and elapsed time compared with those of the conventional optical mark recognition (OMR) systems. The proposed system also demonstrated that it can also achieve high accuracy of 99.7% while using non-transoptic answer sheet paper with lower cost.

SVM classifier based face detection system using BDIP and BVLC moments

In this paper, a support vector machine (SVM) classifier has been used to detect a face in an authentication application. A face candidate is first allocated from the input frame and then normalized to 200×200 pixels images. The textureness of candidates is then measured by the combination of BDIP and BVLC moments and classified into face and non-face ones by a SVM classifier which is known as efficient classification tool. In SVM learning, a DB of 2500 faces and 2500 non-faces has been created under different light conditions and face expressions. The experiments showed that the effectiveness of the used features for SVM based classification issue in the face-detection system.

Simulation and implementation of improved chaotic Colpitts circuit for UWB communications

In this paper, we develop an improved chaotic Colpitts oscillator applied to ultra wide band wireless communication system. We design, simulate and implement the improved chaotic Colpitts oscillator using Philips wideband transistor BJT BFG425W. The numerical simulation using Matlab and circuit simulation using Advanced Design System (ADS) are also presented. The oscillator is designed based on micro-strip substrate and the simulation result of the layout approves the possibility for UWB wireless communication applications.

Detect abnormalities in mammograms by local contrast thresholding and rule-based classification

Mammography, which uses X-ray technology to image the breast, is currently the most effective and reliable method for early cancer detection. There exists limitations of human observers: up 30% of breast lessions are missed during routine screening. It is believed that computer-aided detection (CAD) schemes could ultimately provide a useful “second option” for radiologists and potentially improve their diagnostic accuracy. The proposed detection process bases on local contrast thresholding and rule-based classification which is performed over the preprocessed and segmented mammograms. A relatively high detection rate of suspicious abnormal regions (mass and/or microcalcification) on the testing set of mammograms from Mini Mias Database implies that the proposed method can assist technologists in more efficiently and accurately locating the exact areas for subsequent exams.

An efficient and reliable human resource management system based on a hybrid of face authentication and RFID technology

This paper focuses on the research and development of a human resource management (HRM) system which is based on face authentication and RFID technology for more efficient security and reliability. In the proposed system, a user ID is first read from the given RFID tag to retrieve the corresponding records from the database to be used in the authentication process together with the user image which is captured independently by a camera. Haar like features are exploited to allocate face candidates from the captured images and then classified by a SVM classifier into face and non-face samples, where the textureness and roughness of the samples are measured by multi-window BDIP and multi-window BVLC moments evaluated for window sizes of 2× 2, 3× 3 and 4× 4. The authentication process is then performed by utilization of well-known client specific Fisherface method, which is based on combination of PCA and LDA. The experiment showed that the system has achieved the authentication accuracy about 91.3 % with an increased reality and stability.

Implementation of webcam-based touchscreen

This paper presents implementation of webcam-based touchscreen where the main tools used in this work is image processing techniques. Two cameras are employed for acquiring sequence of images. Touching object is recognized and its location is identified. Experiment is carried out to demonstrate and verify the effectiveness of proposed method.

A Program for Locating Possible Breast Masses on Mammograms

Breast cancer is the second most common type of cancer and also one of the leading cause of cancer death worldwide. Mammography, which uses X-ray technology to image the breast, is currently the method of choice of early detection of breast cancer. This paper introduces a method for detecting suspicious masses on mammograms, which are efficient applicable for circumstances in Vietnam. The method consists of three steps: background subtraction, templace matching and region growing. The method has been tested on the Mammography Image Analysis Society (MIAS) database. A relatively high detection rate and low false positive per image implies that the method can be a quite effective tool in diagnosing breast cancer.

Identifying ovine transcranial acoustic windows

Transcranial Doppler ultrasound (TCD) is used to measure cerebral blood flow velocity in patients with cerebrovascular pathology. The task of locating an adequate acoustic window, through which TCD may be performed, is frequently time consuming or impossible for patients with thick or dense temporal bones. We examined the possibility of utilising B-Mode ultrasound images to guide detection of acoustic windows prior to TCD. B-Mode images of the temporal bone were acquired on deceased sheep shortly after death. We found that on selected sheep, the inner table of the temporal bone was visible on B-Mode imaging. We suggest that visibility of the inner table of the temporal bone on B-Mode imaging is associated with a high probability of successful TCD at that location.

Towards ultrasonic detection of acoustic windows for transcranial Doppler ultrasound and related procedures

Transcranial Ultrasound (TCD) is a neurosonological procedure to examine intracranial brain supplying arteries, by visualizing cerebral blood flow patterns. In patients suffering from severe blood flow disruption to the brain neurological pathology conditions such as stenosis, occlusions and vasospasm can be identified with TCD. Ischemic and hemorrhagic strokes require regular TCD monitoring. However, the TCD procedure is largely dependent on the operators skills and experiences. Finding a suitable site on the temporal plane (an acoustic window) for transcranial insonoification can be challenging in patients with poor or no acoustic windows. In such cases, diagnosis and prognosis may be problematic. Approximately, 10–20% of patients have poor or no acoustic windows. Therefore, early detection of acoustic windows for transcranial sonograph is desirable. This paper examines ultrasound signals in-vitro bovines biological transcranial media to early detect acoustic windows. The results suggest that transcranial acoustic windows can be detected with an ultrasound signal. Ultrasound attenuation-velocity and B-Mode Ultrasound detection contrast the temporal bone and the surrounding soft-tissue, suggesting that acoustic windows can be detected.

Chaotic security based on non-stationary dynamics and random manifolds

In this paper, we propose secure communication schemes using projective-lag and/or projective-anticipating synchronization or multidelay feedback systems and the modulation scheme of synchronization-manifold shift keying. The strategy to assure the security is based on non-stationary dynamics and random synchronization manifolds. Where, random synchronization manifolds are generated to encode various blocks of bits during transmission time. Numerical simulation will confirm the performance of the proposed security schemes.