• 한국컴퓨터정보학회논문지
• /
• 제24권6호
• /
• pp.11-19
• /
• 2019

In this paper, a new Scene-based Nonuniformity Correction (SBNUC) method is proposed by applying Image Roughness-like and Spatial Noise cost functions on deep neural network structure. The classic approaches for nonuniformity correction require generally plenty of sequential image data sets to acquire accurate image correction offset coefficients. The proposed method, however, is able to estimate offset from only a couple of images powered by the characteristic of deep neural network scheme. The real world SWIR image set is applied to verify the performance of proposed method and the result shows that image quality improvement of PSNR 70.3dB (maximum) is achieved. This is about 8.0dB more than the improved IRLMS algorithm which preliminarily requires precise image registration process on consecutive image frames.

• Current Optics and Photonics
• /
• 제8권4호
• /
• pp.391-398
• /
• 2024

An image reconstruction algorithm is vital for the image quality of a photonic integrated interferometric imaging (PIII) system. However, image reconstruction algorithms have limitations that always lead to degraded image reconstruction. In this paper, a novel image reconstruction algorithm based on deep learning is proposed. Firstly, the principle of optical signal transmission through the PIII system is investigated. A dataset suitable for image reconstruction of the PIII system is constructed. Key aspects such as model and loss functions are compared and constructed to solve the problem of image blurring and noise influence. By comparing it with other algorithms, the proposed algorithm is verified to have good reconstruction results not only qualitatively but also quantitatively.

• 한국정보과학회논문지:시스템및이론
• /
• 제30권11호
• /
• pp.621-628
• /
• 2003

본 논문은 한 장의 영상 속에 포함되어 있는 물체들이 외관상의 3차원 운동이 가능하도록 하는 방법을 제시한다. 이전 연구들은 여러 장의 영상으로부터 영상 기반 모델링 기법들을 이용하여 3차원 모델을 생성하거나 소실점을 이용한 카메라 보정을 통하여 장면을 입방체로 모델링하여 3차원 모델을 생성하는 방식으로 접근하였다. 그러나 본 논문에서는 장면의 기하학적 정보나 카메라 보정 없이 장면 속 물체의 영상 기반 운동(image-based motion)의 가능성을 제시한다. 구현된 시스템은 영상을 시점에 관한 사영 평면으로 생각하고 사용자에 의해 입력된 선과 점의 정보를 이용하여 사영된 3차원 물체의 2차원 모양을 모델링한다. 그리고 모델링된 물체는 3차원 운동을 하기 위한 지역 좌표계로서 소실점을 이용한다.

• ETRI Journal
• /
• 제28권1호
• /
• pp.51-58
• /
• 2006

In this paper, we propose a method of applying a lifting-based wavelet domain e-median filter (LBWDEMF) for image restoration. LBWDEMF helps in reducing the number of computations. An e-median filter is a type of modified median filter that processes each pixel of the output of a standard median filter in a binary manner, keeping the output of the median filter unchanged or replacing it with the original pixel value. Binary decision-making is controlled by comparing the absolute difference of the median filter output and the original image to a preset threshold. In addition, the advantage of LBWDEMF is that probabilities of encountering root images are spread over sub-band images, and therefore the e-median filter is unlikely to encounter root images at an early stage of iterations and generates a better result as iteration increases. The proposed method transforms an image into the wavelet domain using lifting-based wavelet filters, then applies an e-median filter in the wavelet domain, transforms the result into the spatial domain, and finally goes through one spatial domain e-median filter to produce the final restored image. Moreover, in order to validate the effectiveness of the proposed method we compare the result obtained using the proposed method to those using a spatial domain median filter (SDMF), spatial domain e-median filter (SDEMF), and wavelet thresholding method. Experimental results show that the proposed method is superior to SDMF, SDEMF, and wavelet thresholding in terms of image restoration.

• 한국멀티미디어학회논문지
• /
• 제17권12호
• /
• pp.1393-1401
• /
• 2014

This paper presents an adaptive image interpolation method using a pixel-based neighbor embedding which is modified from the patch-based neighbor embedding of contemporary super resolution algorithms. Conventional interpolation methods for high resolution detect at least 16-directional edges in order to remove zig-zaging effects and selectively choose the interpolation strategy according to the direction and value of edge. Thus, they require much computation and high complexity. In order to develop a simple interpolation method preserving edge's directional shape, the proposed algorithm adopts the simplest Haar wavelet and suggests a new pixel-based embedding scheme. First, the low-quality image but high resolution, magnified into 1 octave above, is acquired using an adaptive 8-directional interpolation based on the high frequency coefficients of the wavelet transform. Thereafter, the pixel embedding process updates a high resolution pixel of the magnified image with the weighted sum of the best matched pixel value, which is searched at its low resolution image. As the results, the proposed scheme is simple and removes zig-zaging effects without any additional process.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권2호
• /
• pp.837-856
• /
• 2016

Underwater image enhancement has received considerable attention in last decades, due to the nature of poor visibility and low contrast of underwater images. In this paper, we propose a new automatic underwater image enhancement algorithm, which combines nonsubsampled contourlet transform (NSCT) domain enhancement techniques with the mechanism of the human visual system (HVS). We apply the multiscale retinex algorithm based on the HVS into NSCT domain in order to eliminate the non-uniform illumination, and adopt the threshold denoising technique to suppress underwater noise. Our proposed algorithm incorporates the luminance masking and contrast masking characteristics of the HVS into NSCT domain to yield the new HVS-based NSCT. Moreover, we define two nonlinear mapping functions. The first one is used to manipulate the HVS-based NSCT contrast coefficients to enhance the edges. The second one is a gain function which modifies the lowpass subband coefficients to adjust the global dynamic range. As a result, our algorithm can achieve contrast enhancement, image denoising and edge sharpening automatically and simultaneously. Experimental results illustrate that our proposed algorithm has better enhancement performance than state-of-the-art algorithms both in subjective evaluation and quantitative assessment. In addition, our algorithm can automatically achieve underwater image enhancement without any parameter tuning.

• 인터넷정보학회논문지
• /
• 제21권5호
• /
• pp.67-73
• /
• 2020

In this paper, we present a CNN-based gesture recognition approach which reduces the memory burden of input data. Most of the neural network-based gesture recognition methods have used a sequence of frame images as input data, which cause a memory burden problem. We use a motion history image in order to define a meaningful gesture. The motion history image is a grayscale image into which the temporal motion information is collapsed by synthesizing silhouette images of a user during the period of one meaningful gesture. In this paper, we first summarize the previous traditional approaches and neural network-based approaches for gesture recognition. Then we explain the data preprocessing procedure for making the motion history image and the neural network architecture with three convolution layers for recognizing the meaningful gestures. In the experiments, we trained five types of gestures, namely those for charging power, shooting left, shooting right, kicking left, and kicking right. The accuracy of gesture recognition was measured by adjusting the number of filters in each layer in the proposed network. We use a grayscale image with 240 × 320 resolution which defines one meaningful gesture and achieved a gesture recognition accuracy of 98.24%.

• 한국통신학회논문지
• /
• 제34권10B호
• /
• pp.1086-1092
• /
• 2009

Stereoscopic conversion has been an important and challenging issue for many 3-D video applications. Usually, there are two different stereoscopic conversion approaches, i.e., image motion-based conversion that uses motion information and object-based conversion that partitions an image into moving or static foreground object(s) and background and then converts the foreground in a stereoscopic object. As well, since the input sequence is MPEG-1/2 compressed video, motion data stored in compressed bitstream are often unreliable and thus the image motion-based conversion might fail. To solve this problem, we present the utilization of key-motion that has the better accuracy of estimated or extracted motion information. To deal with diverse motion types, a transform space produced from motion vectors and color differences is introduced. A key-motion is determined from the transform space and its associated stereoscopic image is generated. Experimental results validate effectiveness and robustness of the proposed method.

• 한국컴퓨터정보학회논문지
• /
• 제23권4호
• /
• pp.147-153
• /
• 2018

In this paper, we propose an efficient WBC 14-Diff classification which performs using the WBC-ResNet-152, a type of CNN model. The main point of view is to use Super-pixel for the segmentation of the image of WBC, and to use ResNet for the classification of WBC. A total of 136,164 blood image samples (224x224) were grouped for image segmentation, training, training verification, and final test performance analysis. Image segmentation using super-pixels have different number of images for each classes, so weighted average was applied and therefore image segmentation error was low at 7.23%. Using the training data-set for training 50 times, and using soft-max classifier, TPR average of 80.3% for the training set of 8,827 images was achieved. Based on this, using verification data-set of 21,437 images, 14-Diff classification TPR average of normal WBCs were at 93.4% and TPR average of abnormal WBCs were at 83.3%. The result and methodology of this research demonstrates the usefulness of artificial intelligence technology in the blood cell image classification field. WBC-ResNet-152 based morphology approach is shown to be meaningful and worthwhile method. And based on stored medical data, in-depth diagnosis and early detection of curable diseases is expected to improve the quality of treatment.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 1999년도 KOBA 방송기술 워크샵 KOBA Broadcasting Technology Workshop
• /
• pp.133-140
• /
• 1999

In this paper, we propose a method of image-based three-dimensional modeling for realistic facial expression. In the proposed method, real human facial images are used to deform a generic three-dimensional mesh model and the deformed model is animated to generate facial expression animation. First, we take several pictures of the same person from several view angles. Then we project a three-dimensional face model onto the plane of each facial image and match the projected model with each image. The results are combined to generate a deformed three-dimensional model. We use the feature-based image metamorphosis to match the projected models with images. We then create a synthetic image from the two-dimensional images of a specific person's face. This synthetic image is texture-mapped to the cylindrical projection of the three-dimensional model. We also propose a muscle-based animation technique to generate realistic facial expression animations. This method facilitates the control of the animation. lastly, we show the animation results of the six represenative facial expressions.