• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.4 no.2
• /
• pp.371-378
• /
• 2000

In this study, we proposed pathological pattern detection system for X-ray chest image using artificial neural network. In a physical examination, radiologists have checked on the chest image projected the view box by a magnifying glass and found out what the disease is. Here, the detection of X-ray fluoroscopy is tedious and time-consuming for human doing. Lowering of efficiency for chest diagnosis is caused by lots mistakes of radiologist because of detecting the micro pathology from the film of small size. So, we proposed the method for disease detection using artificial neural network and digital image processing on a X-ray chest image. This method composes the function of image sampling, median filter, image equalizer used neural network and pattern recognition used neural network. We confirm this method has improved the problem of a conventional method.

• Korean Journal of Geomatics
• /
• v.1 no.1
• /
• pp.43-49
• /
• 2001

Although many kinds of stereo matching method have been developed in the field of computer vision and photogrammetry, wrong matches are not easy to avoid. This paper presents a new method to reduce wrong matches after matching, and experimental results are reported. The main idea is to analyze the histogram of the image attribute differences between each pair of image patches matched. Typical image attributes of image patch are the mean and the standard deviation of gray value for each image patch, but there could be other kinds of image attributes. Another idea is to check relative position among potential matches. This paper proposes to use Gaussian blunder filter to detect the suspicious pair of candidate match in relative position among neighboring candidate matches. If the suspicious candidate matches in image attribute difference or relative position are suppressed, then many wrong matches are removed, but minimizing the suppression of good matches. The proposed method is easy to implement, and also has potential to be applied as post processing after image matching for many kinds of matching methods such as area based matching, feature matching, relaxation matching, dynamic programming, and multi-channel image matching. Results show that the proposed method produces fewer wrong matches than before.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.6 no.4
• /
• pp.840-866
• /
• 2014

Visibility in underwater images is usually poor because of the attenuation of light in the water that causes low contrast and color variation. In this paper, a new approach for underwater image quality improvement is presented. The proposed method aims to improve underwater image contrast, increase image details, and reduce noise by applying a new method of using contrast stretching to produce two different images with different contrasts. The proposed method integrates the modification of the image histogram in two main color models, RGB and HSV. The histograms of the color channel in the RGB color model are modified and remapped to follow the Rayleigh distribution within certain ranges. The image is then converted to the HSV color model, and the S and V components are modified within a certain limit. Qualitative and quantitative analyses indicate that the proposed method outperforms other state-of-the-art methods in terms of contrast, details, and noise reduction. The image color also shows much improvement.

• Journal of Korea Multimedia Society
• /
• v.11 no.6
• /
• pp.764-776
• /
• 2008

Image identification is the process of checking whether the query image is the transformed version of the specific original image or not. In this paper, image identification method based on feature composition is proposed. Used features include color distance, texture information and average pixel intensity. We extract color characteristics using color distance and texture information by Modified Generalized Symmetry Transform as well as average intensity of each pixel as features. Individual feature is quantized adaptively to be used as bins of histogram. The histogram is normalized according to data type and it is used as the signature in comparing the query image with database images. In matching part, Manhattan distance is used for measuring distance between two signatures. To evaluate the performance of the proposed method, independent test and accuracy test are achieved. In independent test, 60,433 images are used to evaluate the ability of discrimination between different images. And 4,002 original images and its 29 transformed versions are used in accuracy test, which evaluate the ability that the proposed algorithm can find the original image correctly when some transforms was applied in original image. Experiment results show that the proposed identification method has good performance in accuracy test. And the proposed method is very useful in real environment because of its high accuracy and fast matching capacity.

• The KIPS Transactions:PartB
• /
• v.9B no.1
• /
• pp.91-98
• /
• 2002

In this paper, an implementation method of the efficient image transmission stage using watermarking and channel ceding is proposed. Usually, image communication system consists of both a transmitter part and a receiver part. The transmitter part takes charge of copyright protection of the generated image data, and image coding and compression that can deal with channel noises when transmitting. In the transmitter part, we propose a channel coding method which protects both the watermark signal and the original signal for protecting the copyright of image data and solving channel noises when transmitting. Firstly, copyright protection of image data is conducted. For this, image structure analysis is performed, and both the improvement of image quality and the generation of the watermark signal are made. Then, the histogram is constructed and the watermark signals are selected from this. At this stage, by embedding of the coefficients of curve fittness into the lower 4 bits of the image data pixels, image quality degradation due to the embedding of watermark signals are prevented. Finally, turbo coding, which has the most efficient error correction capability in error correction codes, has been conducted to protect signals of watermark and preserved original image quality against noises on the transmission channel. Particularly, a new interleaving method named "semi random inter]easer" has been proposed.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.05a
• /
• pp.657-660
• /
• 2009

An image interpolation is a basis technique for various image processing and is required to minimize approaches for image quality deterioration. In this paper, we propose an improved bilinear interpolation using loss information estimation. In the proposed algorithm, we estimate loss information of low resolution image using down-sampling and interpolation of acquisition low resolution. The estimated loss information is utilized interpolated image, and it decrease image quality deterioration. Our experiments obtained the average PSNR 0.97~1.79dB which is improved results better than conventional method for sensitive image quality. Also, subjective image quality with edge region is more clearness. The proposed method may be helpful for applications in various multimedia systems such as image resolution enhancement and image restoration.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.15 no.2
• /
• pp.60-64
• /
• 2011

Purpose: The brain perfusion SPECT is the examination which is able to know adversity information related brain disorder. But brain perfusion SPECT has also high failure rates by patient's motions. In this case, we have to use two days method and patients put up with many disadvantages. We think that we don't use two days method in brain perfusion SPECT, if we can use registration method. So this study has led to look over registration method applications in brain perfusion SPECT. Materials and Methods: Jaszczak, Hoffman and cylindrical phantoms were used for acquiring SPECT image data on varying degree in x, y, z axes. The phantoms were filled with $^{99m}Tc$ solution that consisted of a radioactive concentration of 111 MBq/mL. Phantom images were acquired through scanning for 5 sec long per frame by using Triad XLT9 triple head gamma camera (TRIONIX, USA). We painted the ROI of registration image in brain data. So we calculated the ROIratio which was different original image counts and registration image counts. Results: When carring out the experiments under the same condition, total counts differential was from 3.5% to 5.7% (mean counts was from 3.4% to 6.8%) in phantom and patients data. In addition, we also run the experiments in the double activity condition. Total counts differential was from 2.6% to 4.9% (mean counts was from 4.1% to 4.9%) in phantom and patients data. Conclusion: We can know that original and registration data are little different in image analysis. If we use the image registration method, we can improve disadvantage of two days method in brain perfusion SPECT. But we must consider image registration about the distance differences in x, y, z axes.

• Journal of Broadcast Engineering
• /
• v.26 no.1
• /
• pp.26-38
• /
• 2021

In this paper, DLKF (Deep Learning Keypoint Filtering), the deep learning-based keypoint filtering method for the rapidization of the image registration method for remote sensing images is proposed. The complexity of the conventional feature-based image registration method arises during the feature matching step. To reduce this complexity, this paper proposes to filter only the keypoints detected in the artificial structure among the keypoints detected in the keypoint detector by ensuring that the feature matching is matched with the keypoints detected in the artificial structure of the image. For reducing the number of keypoints points as preserving essential keypoints, we preserve keypoints adjacent to the boundaries of the artificial structure, and use reduced images, and crop image patches overlapping to eliminate noise from the patch boundary as a result of the image segmentation method. the proposed method improves the speed and accuracy of registration. To verify the performance of DLKF, the speed and accuracy of the conventional keypoints extraction method were compared using the remote sensing image of KOMPSAT-3 satellite. Based on the SIFT-based registration method, which is commonly used in households, the SURF-based registration method, which improved the speed of the SIFT method, improved the speed by 2.6 times while reducing the number of keypoints by about 18%, but the accuracy decreased from 3.42 to 5.43. Became. However, when the proposed method, DLKF, was used, the number of keypoints was reduced by about 82%, improving the speed by about 20.5 times, while reducing the accuracy to 4.51.

• Smart Structures and Systems
• /
• v.24 no.5
• /
• pp.659-668
• /
• 2019

In digital image correlation, speckle image is closely related to the measurement accuracy. A practical global evaluation criterion for speckle image is presented. Firstly, based on the essential factors of the texture image, both the average particle size and image sharpness are used for the assessment of speckle image. The former is calculated by a simplified auto-covariance function and Gaussian fitting, and the latter by focusing function. Secondly, the computation of the average particle size and image sharpness is verified by numerical simulation. The influence of these two evaluation parameters on mean deviation and standard deviation is discussed. Then, a physical model from speckle projection to image acquisition is established. The two evaluation parameters can be mapped to the physical devices, which demonstrate that the proposed evaluation method is reasonable. Finally, the engineering application of the evaluation method is pointed out.

• Spatial Information Research
• /
• v.18 no.3
• /
• pp.73-83
• /
• 2010

This study presents a method to segment an image, employing the statistics observed at each pixel location across sequential frame images. In the acquisition and analysis of spatial information, utilization of digital image processing technique has very important implications. Various image segmentation techniques have been presented to distinguish the area of digital images. In this study, based on the analysis of the spectroscopic characteristics of sequential frame images that had been previously researched, an image segmentation method was proposed by using the randomness occurring among a sequence of frame images for a same scene. First of all, we computed the mean and standard deviation values at each pixel and found reliable pixels to determine seed points using their standard deviation value. For segmenting an image into individual regions, we conducted region growing based on a T-test between reference and candidate sample sets. A comparative analysis was conducted to assure the performance of the proposed method with reference to a previous method. From a set of experimental results, it is confirmed that the proposed method using a sequence of frame images segments a scene better than a method using a single frame image.