Younghun Sung

Highly stable temperature characteristics of InGaN blue laser diodes

We report stable temperature characteristics of threshold current and output power in InGaN blue laser diodes emitting around 450nm. The threshold current is changed by <3mA in operation temperature range from 20to80°C, and even negative characteristic temperature is observed in a certain temperature range. This peculiar temperature characteristic is attributed to originate from unique carrier transport properties of InGaN quantum wells with high In composition, which is deduced from the simulation of carrier density and optical gain. In addition, slope efficiency is also maintained well and wall plug efficiency is even improved as temperature increases.

Recent achievements of AlInGaN based laser diodes in blue and green wavelength

AlInGaN based blue and blue-green LDs were investigated with regard to the characteristics of GaN semiconductor laser diodes. High power, single mode blue LDs with high COD level (~334mW under CW operation at 25°C, kink-free at 150mW) and long lifetime (~10000 hours under CW operation, 50mW 25°C) were achieved. No significant characteristic differences between blue LDs on LEO-GaN/sapphire and GaN substrate were observed. The blue-green LD which has the wavelength of 485 nm was successfully fabricated and demonstrated under CW operation 25°C, while it showed poor performances of LD characteristics compared to those of blue LDs. We believe that the poor performance of blue-green LDs were caused by the piezo-electric effect by lattice mismatch along C-axis of GaN, In fluctuation by lattice mismatch and In solubility limit in InGaN QWs and thermal annealing which was performed during the p-layer growth.

An Asynchronous Sampling-Based 128

This paper presents a direct photon-counting X-ray image detector with a HgI2 photoconductor for high-quality medical imaging applications. The proposed sampling-based charge preamplifier with asynchronous self-reset enables a pixel to detect single X-ray photon energy with higher sensitivity and faster processing rate. The use of the correlated double sampling enabled by the sampling-based architecture also reduces flicker noise and contributes to the achievement of high pixel-to-pixel uniformity. Discrimination of the energy level of the detected X-rays is performed by the proposed compact in-pixel ADC with low power consumption. Three 15-bit counters in each pixel count up energy-discriminated photons for the reconstruction of multispectral X-ray images. A 128 × 128 X-ray image detector with a pixel size of 60 × 60 μm2 is implemented and measured using a 0.13-μm/0.35-μm standard CMOS process. It discriminates 3 energy levels of photon energy with a gain of 107 mV/ke- and a static power consumption of 4.6 μW/pixel. The measured equivalent noise charge (ENC) and minimum detectable energy level of the detector pixel are 68 e- rms and 290 e-, respectively. The measured maximum threshold dispersion in the pixel array is 164 e- rms without any calibration. The functionality of our chip is also successfully demonstrated using real X-ray images.

\times

We present an example-based learning approach for detecting a partially occluded human face in a scene provided by a camera of Automated Teller Machine (ATM) in a bank. Gradient mapping in scale space is applied on an original image, providing human face representation robust to illumination variance. Detection of the partially occluded face, which can be used in characterization of suspicious ATM users, is then performed based on Support Vector Machine (SVM) method. Experimental results show that a high detection rate over 95% is achieved in image samples acquired from in-use ATM.

128 Direct Photon-Counting X-Ray Image Detector with Multi-Energy Discrimination and High Spatial Resolution

High power and high efficiency AlInGaN-based laser diodes with 405 nm were fabricated for the post-DVD applications. Magnesium doped AlGaN/GaN multiple quantum barrier (MQB) layers were introduced into the laser diode structure, which resulted in considerable improvement in lasing performances such as threshold current and slope efficiency. Asymmetric waveguide structure was used in order to improve the characteristics of laser diodes. Aluminum content in the n-cladding layer was varied in connection with the vertical beam divergence angle and COD level. By decreasing Al content in the n-cladding layer, the vertical divergence angle was reduced to 17 degree and the COD level was enhanced to over 300mW. The maximum output power reached as high as 470 mW, the highest value ever reported for the narrow-stripe GaN LDs. In addition, the fundamental transverse-mode operation was clearly demonstrated up to 500 mW-pulsed output power.

A new video surveillance system employing occluded face detection

X-ray projection is not effective for representing complex overlapping objects. This paper presents a novel computational framework to decompose X-ray projections into multiple images with non-overlapping objects that are differentiated by their own material compositions. Based on energy-dependent X-ray attenuation characteristics for each material, multiple energy X-ray images are analyzed to obtain material-selective images, which correspond to projections of basis materials that constitute objects. We show that material-selective images can be considered as linear mixtures of independent components that are associated with object-selective images. As a result, multiple objects can be decomposed by independent component analysis (ICA) of material-selective images or ICA of multiple monochromatic energy X-ray images. To demonstrate the concept of the proposed method, we apply it to simulated images based on a 3-D human model.

High power AlInGaN-based blue-violet laser diodes

Accurate and efficient scatter correction is essential for acquisition of high-quality x-ray cone-beam CT (CBCT) images for various applications. This study was conducted to demonstrate the feasibility of using the data consistency condition (DCC) as a criterion for scatter kernel optimization in scatter deconvolution methods in CBCT. As in CBCT, data consistency in the mid-plane is primarily challenged by scatter, we utilized data consistency to confirm the degree of scatter correction and to steer the update in iterative kernel optimization. By means of the parallel-beam DCC via fan-parallel rebinning, we iteratively optimized the scatter kernel parameters, using a particle swarm optimization algorithm for its computational efficiency and excellent convergence. The proposed method was validated by a simulation study using the XCAT numerical phantom and also by experimental studies using the ACS head phantom and the pelvic part of the Rando phantom. The results showed that the proposed method can effectively improve the accuracy of deconvolution-based scatter correction. Quantitative assessments of image quality parameters such as contrast and structure similarity (SSIM) revealed that the optimally selected scatter kernel improves the contrast of scatter-free images by up to 99.5%, 94.4%, and 84.4%, and of the SSIM in an XCAT study, an ACS head phantom study, and a pelvis phantom study by up to 96.7%, 90.5%, and 87.8%, respectively. The proposed method can achieve accurate and efficient scatter correction from a single cone-beam scan without need of any auxiliary hardware or additional experimentation.

Multiple object decomposition based on independent component analysis of multi-energy x-ray projections

In this paper, a novel reconstruction algorithm for limited angle tomography using total variation (TV) regularization is presented. Inspired by duality-based TV minimization in denoising and deblurring applications, we derived a TV regularized statistical reconstruction algorithm composed of relatively simple and structured operations such as discrete gradient and divergence calculations, which presents an effective way to introduce TV regularization to the statistical reconstruction. In initial tests with real data from a digital breast tomosynthesis system, the proposed algorithm showed reliable reconstructions for low dose conditions.

Data consistency-driven scatter kernel optimization for x-ray cone-beam CT

PURPOSE Material decomposition using multienergy photon counting x-ray detectors (PCXD) has been an active research area over the past few years. Even with some success, the problem of optimal energy selection and three material decomposition including malignant tissue is still on going research topic, and more systematic studies are required. This paper aims to address this in a unified statistical framework in a mammographic environment. METHODS A unified statistical framework for energy level optimization and decomposition of three materials is proposed. In particular, an energy level optimization algorithm is derived using the theory of the minimum variance unbiased estimator, and an iterative algorithm is proposed for material composition as well as system parameter estimation under the unified statistical estimation framework. To verify the performance of the proposed algorithm, the authors performed simulation studies as well as real experiments using physical breast phantom and ex vivo breast specimen. Quantitative comparisons using various performance measures were conducted, and qualitative performance evaluations for ex vivo breast specimen were also performed by comparing the ground-truth malignant tissue areas identified by radiologists. RESULTS Both simulation and real experiments confirmed that the optimized energy bins by the proposed method allow better material decomposition quality. Moreover, for the specimen thickness estimation errors up to 2 mm, the proposed method provides good reconstruction results in both simulation and real ex vivo breast phantom experiments compared to existing methods. CONCLUSIONS The proposed statistical framework of PCXD has been successfully applied for the energy optimization and decomposition of three material in a mammographic environment. Experimental results using the physical breast phantom and ex vivo specimen support the practicality of the proposed algorithm.

Statistical reconstruction using dual formulation of subband-wise total variation regularization (SDST) for limited angle tomography

The X-ray tomosynthesis that measures several low dose projections over a limited angular range has been investigated as an alternative method of X-ray mammography for breast cancer screening. An extension of the scan coverage increases the vertical resolution by mitigating the interplane blurring. The implementation of a wide angle tomosynthesis equipment, however, may not be straightforward, mainly due to the image deterioration from the statistical noise in exterior projections. In this paper, we adopt the voltage modulation scheme to enlarge the coverage of the tomosynthesis scan. The higher tube voltages are used for outer angles, which offers the sufficient penetrating power for outlying frames in which the pathway of X-ray photons is elongated. To reconstruct 3D information from voltage modulated projections, we propose a novel algorithm, named information theoretic discrepancy based iterative reconstruction (IDIR) algorithm, which allows to account for the polychromatic acquisition model. The generalized information theoretic discrepancy (GID) is newly employed as the objective function. Using particular features of the GID, the cost function is derived in terms of imaginary variables with energy dependency, which leads to a tractable optimization problem without using the monochromatic approximation. In preliminary experiments using simulated and experimental equipment, the proposed imaging architecture and IDIR algorithm showed superior performances over conventional approaches.