• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.733-735
• /
• 2014

Example based Super Resolution(SR) is using the correspondence between the low and high resolution image from a database. This method uses only one image to estimate a high resolution image and can get the larger image than 2 times. Example based SR is proposed to solve the problem of classical SR. Neighbor embedding(NE) has been inspired by manifold learning method, particularly locally linear embedding. However, the poor generalization of NE decreases the performance of such algorithm. The sizes of local training sets are always too small to improve the performance of NE. We propose the advanced NE baesd on SVR having an excellent generalization ability to solve this problem. Given a low resolution image, we estimate a pixel in its high resolution version by using SVR based NE. Through experimental results, we quantitatively and qualitatively confirm the improved results of the proposed algorithm when comparing with conventional interpolation methods and NE.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.12
• /
• pp.69-77
• /
• 2017

This paper represents a novel framework for multi-scale image fusion. Multi-scale Kalman Smoothing (MKS) algorithm with quad-tree structure can provide a powerful multi-resolution image fusion scheme by employing Markov property. In general, such approach provides outstanding image fusion performance in terms of accuracy and efficiency, however, quad-tree based method is often limited to be applied in certain applications due to its stair-like covariance structure, resulting in unrealistic blocky artifacts at the fusion result where finest scale data are void or missed. To mitigate this structural artifact, in this paper, a new scheme of multi-scale fusion framework is proposed. By employing Super Resolution (SR) technique on MKS algorithm, fine resolved measurement is generated and blended through the tree structure such that missed detail information at data missing region in fine scale image is properly inferred and the blocky artifact can be successfully suppressed at fusion result. Simulation results show that the proposed method provides significantly improved fusion results in the senses of both Root Mean Square Error (RMSE) performance and visual improvement over conventional MKS algorithm.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1486-1494
• /
• 2021

Character recognition is a technology required in various platforms, such as smart parking and text to speech, and many studies are being conducted to improve its performance through new attempts. However, with low-quality image used for character recognition, a difference in resolution of the training image and test image for character recognition occurs, resulting in poor accuracy. To solve this problem, this paper designed an end-to-end learning neural network that combines image super-resolution and character recognition so that the character recognition model performance is robust against various quality data, and implemented an alternative whole learning algorithm to learn the whole neural network. An alternative end-to-end learning and recognition performance test was conducted using the license plate image among various text images, and the effectiveness of the proposed algorithm was verified with the performance test.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.48 no.9
• /
• pp.22-30
• /
• 2011

Recently, the rising demand for intelligent video surveillance system leads to high-performance face recognition systems. The solution for low-resolution images acquired by a long-distance camera is required to overcome the distance limits of the existing face recognition systems. For that reason, this paper proposes a hardware design of an image resolution enhancement algorithm for real-time intelligent video surveillance systems. The algorithm is synthesizing a high-resolution face image from an input low-resolution image, with the help of a large collection of other high-resolution face images, called training set. When we checked the performance of the algorithm at 32bit RISC micro-processor, the entire operation took about 25 sec, which is inappropriate for real-time target applications. Based on the result, we implemented the hardware module and verified it using Xilinx Virtex-4 and ARM9-based embedded processor(S3C2440A). The designed hardware can complete the whole operation within 33 msec, so it can deal with 30 frames per second. We expect that the proposed hardware could be one of the solutions not only for real-time processing at the embedded environment, but also for an easy integration with existing face recognition system.

• Journal of Broadcast Engineering
• /
• v.21 no.2
• /
• pp.180-191
• /
• 2016

When up-scaling algorithms for visible light images are applied to infrared (IR) images, they rarely work because IR images are usually blurred. In order to solve such a problem, this paper proposes an up-scaling algorithm for IR images. We employ adaptive dynamic range encoding (ADRC) as a simple classifier based on the observation that IR images have weak details. Also, since human visual systems are more sensitive to edges, our algorithm focuses on edges. Then, we add pre-processing in learning phase. As a result, we can improve visibility of IR images without increasing computational cost. Comparing with Anchored neighborhood regression (A+), the proposed algorithm provides better results. In terms of just noticeable blur, the proposed algorithm shows higher values by 0.0201 than the A+, respectively.

• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.293-301
• /
• 2022

Direction of arrival (DOA) estimation of space signals is a basic problem in array signal processing. DOA estimation based on the multiple signal classification (MUSIC) algorithm can theoretically overcome the Rayleigh limit and achieve super resolution. However, owing to its inadequate real-time performance and accuracy in practical engineering applications, its applications are limited. To address this problem, in this study, a DOA estimation algorithm with high parallelism and precision based on an analysis of the characteristics of complex matrix eigenvalue decomposition and the coordinate rotation digital computer (CORDIC) algorithm is proposed. For parallel and single precision, floating-point numbers are used to construct an orthogonal identity matrix. Thus, the efficiency and accuracy of the algorithm are guaranteed. Furthermore, the accuracy and computation of the fixed-point algorithm, double-precision floating-point algorithm, and proposed algorithm are compared. Without increasing complexity, the proposed algorithm can achieve remarkably higher accuracy and efficiency than the fixed-point algorithm and double-precision floating-point calculations, respectively.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.12
• /
• pp.1816-1825
• /
• 2022

In this paper, we propose a new approach for tracking each human in a surveillance camera network with various resolution cameras. When tracking human on multiple non-overlapping cameras, the traditional appearance features are easily affected by various camera viewing conditions. To overcome this limitation, the proposed system utilizes facial information along with appearance information. In general, human images captured by the surveillance camera are often low resolution, so it is necessary to be able to extract useful features even from low-resolution faces to facilitate tracking. In the proposed tracking scheme, texture-based face descriptor is exploited to extract features from detected face after face frontalization. In addition, when the size of the face captured by the surveillance camera is very small, a super-resolution technique that enlarges the face is also exploited. The experimental results on the public benchmark Dana36 dataset show promising performance of the proposed algorithm.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.11
• /
• pp.73-81
• /
• 2016

In general, infrared images have less sharpness and image details than visible images. So, the prior image upscaling methods are not effective in the infrared images. In order to solve this problem, this paper proposes an algorithm which initially up-scales an input infrared (IR) image by using adaptive dynamic range encoding (ADRC)-based super-resolution (SR) method, and then fuses the result with the corresponding visible images. The proposed algorithm consists of a up-scaling phase and a fusion phase. First, an input IR image is up-scaled by the proposed ADRC-based SR algorithm. In the dictionary learning stage of this up-scaling phase, so-called 'pre-emphasis' processing is applied to training-purpose high-resolution images, hence better sharpness is achieved. In the following fusion phase, high-frequency information is extracted from the visible image corresponding to the IR image, and it is adaptively weighted according to the complexity of the IR image. Finally, a up-scaled IR image is obtained by adding the processed high-frequency information to the up-scaled IR image. The experimental results show than the proposed algorithm provides better results than the state-of-the-art SR, i.e., anchored neighborhood regression (A+) algorithm. For example, in terms of just noticeable blur (JNB), the proposed algorithm shows higher value by 0.2184 than the A+. Also, the proposed algorithm outperforms the previous works even in terms of subjective visual quality.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.109-111
• /
• 2021

Recently, computer vision research using light field cameras has been actively conducted. Since light field cameras have spatial information, various studies are being conducted in fields such as depth map estimation, super resolution, and 3D object detection. In this paper, we propose a method for detecting objects in blur images through a 7×7 array of images acquired through a light field camera. The blur image, which is weak in the existing camera, is detected through the light field camera. The proposed method uses the SSD algorithm to evaluate the performance using blur images acquired from light field cameras.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.683-684
• /
• 2021

논문은 재구성에 기반을 둔 초 해상도 알고리즘 모델에서 SRCNN과 VDSR의 전반에 걸쳐 구조와 성능에 대하여 알아본다. SRCNN 모델과 VDSR 모델의 구조와 각 방법의 알고리즘 프로세스를 간단히 소개하고 성능 평가 함수의 개선에 대하여 알아보도록 한다.