• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.1
• /
• pp.22-27
• /
• 2014

Blocking and ringing artifacts continue to be the most serious defects that occur in images and video streams compressed to low bit rates using block-based discrete cosine transform(DCT) compression standards. These artifacts contain the high frequency components near the block and the edge boundaries. Usually the lowpass filter can remove them. However, simple lowpass filter results into blur by removing important information such as edges at the same time. To overcome these problems, we propose a novel postfiltering algorithm that calculate the weight value based on the intensity similarity in the neighboring pixels and multiply this weight to the Gaussian lowpass filter coefficient. Experimental results show that the proposed technique provides satisfactory performance in both objective and subjective image quality.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.6
• /
• pp.123-131
• /
• 2009

The emerging display technology has been replaced the previous position of the CRT with the LCD. The nature of hold type display such as LCD, however, causes many problems such as motion blur and motion judder. To resolve the problems, we used frame interpolation technique which improves the image quality by inserting new interpolated frames between existing frames. In this paper, we propose a novel frame interpolation technique that uses dominant MV and variance different value in each block. At first, the proposed algorithm performs unidirectional motion estimation using blocking matching algorithm. The new frame is generated by pixel average using compared block variance or by pixel motion compensation using dominant motion vector, whether the motion estimation find the target area or not. Several experiments with the proposed algorithm shows that the proposed algorithm has better image quality than the existing bidirectional frame interpolation algorithm at the rate of about 3dB PSNR and has low complexity comparing to the unidirectional frame interpolation technique.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.7
• /
• pp.81-89
• /
• 1999

Image obtained by incomplete communication always include noise, blur and distortion, etc. In this paper, we propose and apply the new spatial filter algorithm, called an optimal adaptive generalized stack filter(AGSF), which optimizes adaptive generalized stack filter(AGSF) using neural network weight learning algorithm of back-propagation learning algorithm for improving noise removal and edge preservation rate. AGSF divides into two parts: generalized stack filter(GSF) and adaptive multistage median filter(AMMF), GSF improves the ability of stack filter algorithm and AMMF proposes the improved algorithm for reserving the sharp edge. Applied to neural network theory, the proposed algorithm improves the performance of the AGSF using two weight learning algorithms, such as the least mean absolute(LAM) and least mean square (LMS) algorithms. Simulation results of the proposed filter algorithm are presented and discussed.

• The KIPS Transactions:PartB
• /
• v.16B no.1
• /
• pp.1-6
• /
• 2009

It is very difficult to restore the images degraded by motion blur and additive noise. In conventional methods, regularization usually applies to all the images without considering local characteristics of the images. As a result, ringing artifacts appear in the edge regions and noise amplification is in the flat regions, as well. To solve these problems, we propose an adaptive iterative regularization method, using the way of regularization operator considering edge directions. In addition, we suggest an adaptive regularization parameter and an relaxation parameter. In conclusion, We have verified that the new method shows the suppression of the noise amplification in the flat regions, also does less ringing artifacts in the edge regions. Furthermore, it offers better images and improves the quality of ISNR, comparing with those of conventional methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2390-2406
• /
• 2022

Scene text recognition has important application value and attracted the interest of plenty of researchers. At present, many methods have achieved good results, but most of the existing approaches attempt to improve the performance of scene text recognition from the image level. They have a good effect on reading regular scene texts. However, there are still many obstacles to recognizing text on low-quality images such as curved, occlusion, and blur. This exacerbates the difficulty of feature extraction because the image quality is uneven. In addition, the results of model testing are highly dependent on training data, so there is still room for improvement in scene text recognition methods. In this work, we present a natural scene text recognizer to improve the recognition performance from the feature level, which contains feature representation and feature enhancement. In terms of feature representation, we propose an efficient feature extractor combined with Representative Batch Normalization and ResNet. It reduces the dependence of the model on training data and improves the feature representation ability of different instances. In terms of feature enhancement, we use a feature enhancement network to expand the receptive field of feature maps, so that feature maps contain rich feature information. Enhanced feature representation capability helps to improve the recognition performance of the model. We conducted experiments on 7 benchmarks, which shows that this method is highly competitive in recognizing both regular and irregular texts. The method achieved top1 recognition accuracy on four benchmarks of IC03, IC13, IC15, and SVTP.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.11
• /
• pp.1637-1643
• /
• 2022

In recent years, the demand for image-processing technology in digital marketing has increased due to the expansion and diversification of the digital market, such as video, security, and machine intelligence. Noise-processing is essential for image-correction and reconstruction, especially in the case of sensitive noises, such as in CT, MRI, X-ray, and scanners. The two main salt and pepper noises have been actively studied, but the details and edges are still unsatisfactory and tend to blur when there is a lot of noise. Therefore, this paper proposes an algorithm that applies a weight-based Euclidean distance equation to the partial mask and uses only the non-noisy pixels that are the most similar to the original as effective pixels. The proposed algorithm determines the type of filter based on the state of the internal pixels of the designed partial mask and the degree of mask deterioration, which results in superior noise cancellation even in highly damaged environments.

• Journal of the Korean Society of Radiology
• /
• v.16 no.6
• /
• pp.673-680
• /
• 2022

In this study, based on PLA, we analyzed the Hounsfield Unit (HU) of materials containing 20% each of aluminum, wood, copper, carbon, and marble, and tried to analyze how they affect the image. A cylindrical phantom of 5×30×30 ㎣ (thickness×diameter×height) was fabricated using a entry-level 3D printer. The kV was changed to 80, 100 and 120, and the mAs was changed to 100 and 200 mAs, and the phantom in the center of the table was cross-scanned under a total of six conditions. A circular ROI was set using image J program and the quantification value of the material part HU and the quantification value of the peripheral part CNR were obtained. The HU average of the material part increased in the order of [PLA - wood 20%], [PLA - marble 20%], [PLA - carbon 20%], [PLA 100%], [PLA - aluminum 20%], [PLA - copper 20%] (p<0.05) a negative correlation was confirmed with the HU by increasing kV. It was confirmed that the CNR value in the peripheral area increased in the order of [PLA - marble 20%], [PLA - copper 20%], [PLA - carbon 20%], [PLA - wood 20%], [PLA - aluminum 20%], and [PLA - 100%] (p<0.05). Human organs with similar HU values for each material are [PLA - copper 20%] compact bone, [PLA - aluminum 20%] cancellous bone, [PLA 100%] coagulated blood, [PLA - carbon 20%] and [PLA - marble 20%] liver, muscle, spleen and [PLA - wood 20%] had similar values to fat. In addition, we confirmed the blur phenomenon that blurs the image around the filament with all materials, and confirmed that [PLA 100%] especially has the most blur around the filament. Therefore, it is considered desirable to reflect the HU value of the target organ and consider cloudiness around the phantom when selecting materials for medical phantom fabrication, and this research can provide basic data.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.4
• /
• pp.1-9
• /
• 2012

LCD(Liquid Crystal Display) commonly used as an output device has a drawback of slow response time compared with CRT display. This drawback causes motion blur especially when an abrupt intensity change occurs in an image sequence as time goes on. To overcome the problem of slow response time overdriving technique has been used in TCON of LCD. In this technique, the previous frame data has to be compressed and stored in an external memory. Considering both chip size of TCON and computational complexity, AM-BTC has been applied to the 8bit HD display driver. However, the conventional method is not suitable for 10 bit Full HD because 10 bit Full HD data is much larger than that of 8 bit HD data. Being applied to 10 bit Full HD display driver, the conventional method increase cost by enlarging the external memory size of TCON or deteriorates image quality. In this paper, we propose dual BTC image coding technique for Full HD display driver that is an adaptive coding scheme according to morphological information of each sample block. Through experiments, it is verified that the proposed Dual BTC method performs better than the conventional method not only quantitatively but also qualitatively.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.4
• /
• pp.173-179
• /
• 2014

Today, a surgery for cataract of Korean society frequently come to operation. A method for treating cataracts have been developed in various ways. Widely used method is a method to use the artificial lens and replace it with the existing lens. The surgery can be inserted exactly according to the angle and points that are calculated in advance in the intraocular lens insertion is important. However, The lens insertion point can delete or blur the display due to such factors as foreign material coming out of the eye during surgery. Therefore, The lens insertion point needs a method of tracking the image processing by receiving the camera images in real time display method. In this paper, we propose a feature point tracking method by using template matching and buffer. a simulation results show that our ideas can track a feature point of the intraocular lens insertion.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.12B
• /
• pp.2369-2381
• /
• 1999

In this paper, we propose an iterative algorithm for reducing the blocking artifact in block transform-coded images by using a wavelet transform. In the proposed method, an image is considered as a set of one-dimensional horizontal and vertical signals and one-dimensional wavelet transform is utilized in which the mother wavelet is the first order derivative of a Gaussian like function. The blocking artifact is reduced by removing the blocking component, that causes the variance at the block boundary position in the first scale wavelet domain to be abnormally higher than those at the other positions, using a minimum mean square error (MMSE) filter in the wavelet domain. This filter minimizes the MSE between the ideal blocking component-free signal and the restored signal in the neighborhood of block boundaries in the wavelet domain. It also uses local variance in the wavelet domain for pixel adaptive processing. The filtering and the projection onto a convex set of quantization constraint are iteratively performed in alternating fashion. Experimental results show that the proposed method yields not only a PSNR improvement of about 0.56-1.07 dB, but also subjective quality nearly free of the blocking artifact and edge blur.