• International Journal of Control, Automation, and Systems
• /
• v.1 no.3
• /
• pp.332-338
• /
• 2003

Content-based image retrieval is the research of images from database, that are visually similar to given image examples. Gabor functions and Gabor filters are regarded as excellent methods for feature extraction and texture segmentation. However, they have a disadvantage not to perform well in case of a rotated image because of its direction-oriented filter. This paper proposes a method of extracting local texture features from blocks with central interest points detected in an image and a rotation invariant Gabor wavelet filter. We also propose a method of comparing pattern histograms of features classified by VQ (Vector Quantization) among images.

• Current Optics and Photonics
• /
• v.7 no.3
• /
• pp.237-247
• /
• 2023

In middle- and long-distance imaging systems, due to the atmospheric turbulence caused by temperature, wind speed, humidity, and so on, light waves propagating in the air are distorted, resulting in image-quality degradation such as geometric deformation and fuzziness. In remote sensing, astronomical observation, and traffic monitoring, image information loss due to degradation causes huge losses, so effective restoration of degraded images is very important. To restore images degraded by atmospheric turbulence, an image-restoration method based on improved compound multibranch feature fusion (CMFNetPro) was proposed. Based on the CMFNet network, an efficient channel-attention mechanism was used to replace the channel-attention mechanism to improve image quality and network efficiency. In the experiment, two-dimensional random distortion vector fields were used to construct two turbulent datasets with different degrees of distortion, based on the Google Landmarks Dataset v2 dataset. The experimental results showed that compared to the CMFNet, DeblurGAN-v2, and MIMO-UNet models, the proposed CMFNetPro network achieves better performance in both quality and training cost of turbulent-image restoration. In the mixed training, CMFNetPro was 1.2391 dB (weak turbulence), 0.8602 dB (strong turbulence) respectively higher in terms of peak signal-to-noise ratio and 0.0015 (weak turbulence), 0.0136 (strong turbulence) respectively higher in terms of structure similarity compared to CMFNet. CMFNetPro was 14.4 hours faster compared to the CMFNet. This provides a feasible scheme for turbulent-image restoration based on deep learning.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.11 no.3
• /
• pp.229-234
• /
• 2018

Texture is one of the most useful features in classifying and segmenting images. The LBP-based approach previously presented in the literature has been successful in many applications. However, it's theoretical foundation is based only on the difference of pixel values, and consequently it has a number of drawbacks like it performs poorly for the images corrupted with noise, and especially it cannot be used as a multiscale texture descriptor due to the exploding increase of feature vector dimension with increase of the number of neighbor pixels. In this paper, we present a method to address these drawbacks of LBP-based approach. More specifically, our approach quantizes the range of pixels values and construct a 3D histogram which captures the correlative information of pixels. This histogram is used as a texture feature. Several tests with texture images show that the proposed method outperforms the LBP-based approach in the problem of texture classification.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.13 no.1
• /
• pp.17-24
• /
• 2018

Military camera equipment has a problem that observability is inferior due to various shaking factors. In this paper, we propose an image stabilization algorithm considering performance and execution time to solve this problem and implemented it in Zynq SoC. We stabilized both the simple shaking in the fixed observation position and the sudden shaking in the moving observation position. The feature of the input image is extracted by the Sobel edge algorithm, the subblock with the large edge data is selected, and the motion vector, which is the compensation reference, is calculated through template matching using the 3-step search algorithm of the region of interest. In addition, the proposed algorithm can distinguish the shaking caused by the simple shaking and the movement by using the Kalman filter, and the stabilized image can be obtained by minimizing the loss of image information. To demonstrate the effectiveness of the proposed algorithm, experiments on various images were performed. In comparison, PSNR is improved in the range of 2.6725~3.1629 (dB) and image loss is reduced from 41% to 15%. On the other hand, we implemented the hardware-software integrated design using HLS of Xilinx SDSoC tool and confirmed that it operates at 32 fps on the Zynq board, and realized SoC that operates with real-time processing.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.2
• /
• pp.50-59
• /
• 1994

This paper presents a speech recognition method utilizing third-order cumulants as a feature vector and a neural network for recognition. The use of higher-order cumulants provides desirable uncoupling between the gaussian noise and speech, which enables us to estimate the coefficients of AR model without bias. Unlike the conventional method using second-order statistics, the proposed one exhibits low bias even in SNR as low as 0 dB at the expense of higher variance. It is confirmed through computer simulation that recognition rate of korean single-vowels with the cumulant-based method is much higher than the results with the conventional method even in low SNR.

• Journal of Power Electronics
• /
• v.11 no.3
• /
• pp.360-368
• /
• 2011

This paper proposes an active frequency with a positive feedback in the d-q frame anti-islanding method suitable for a robust phase-locked loop (PLL) algorithm using the FFT concept. In general, PLL algorithms for grid-connected PV PCS use d-q transformation and controllers to make zero an imaginary part of the transformed voltage vector. In a real grid system, the grid voltage is not ideal. It may be unbalanced, noisy and have many harmonics. For these reasons, the d-q transformed components do not have a pure DC component. The controller tuning of a PLL algorithm is difficult. The proposed PLL algorithm using the FFT concept can use the strong noise cancelation characteristics of a FFT algorithm without a PI controller. Therefore, the proposed PLL algorithm has no gain-tuning of a PI controller, and it is hardly influenced by voltage drops, phase step changes and harmonics. Islanding prediction is a necessary feature of inverter-based photovoltaic (PV) systems in order to meet the stringent standard requirements for interconnection with an electrical grid. Both passive and active anti-islanding methods exist. Typically, active methods modify a given parameter, which also affects the shape and quality of the grid injected current. In this paper, the active anti-islanding algorithm for a grid-connected PV PCS uses positive feedback control in the d-q frame. The proposed PLL and anti-islanding algorithm are implemented for a 250kW PV PCS. This system has four DC/DC converters each with a 25kW power rating. This is only one-third of the total system power. The experimental results show that the proposed PLL, anti-islanding method and topology demonstrate good performance in a 250kW PV PCS.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.3
• /
• pp.61-69
• /
• 2018

In this paper, we propose a method for recognizing ground target using dual polarization channels in millimeter waveband air-to-surface radar. First, the Push-Broom target detection method is described and the received signal is modeled considering flight-path scenario of air-to-surface radar. The scattering centers were extracted using the RELAX algorithm, which is a time domain spectral estimation technique, and the feature vector of the target was generated. Based on this, a DB for 4 targets is constructed. As a result of the proposed method, it is confirmed that the target classification rates is improved by more than 15% than the single channel using the data of the dual polarization channel.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.6
• /
• pp.716-721
• /
• 2003

This paper deals with the iris recognition as one of biometric techniques which are applied to identify a person using his/her behavior or congenital characteristics. The iris of a human eye has a texture that is unique and time invariant for each individual. First, we obtain the feature vector from the 2D iris pattern having a property of size invariant and divide it into 24 sectors which are further through three types of 2D Gabor filters. At the recognition process, we compute the similarity measure based on the correlation values. Here, since we use three different matching values obtained from three different directional Gabor filters and select the maximum value among them, it is possible to minimize the recognition error rate. To show the usefulness of the proposed algorithm, we applied it to a biometric database consisting of 50 iris patterns extracted from 10 subjects and finally get more higher than 90% recognition rate.

• Journal of KIISE:Software and Applications
• /
• v.26 no.10
• /
• pp.1143-1155
• /
• 1999

셀룰러 방식의 이동통신 시스템에서 전파의 유효신호 도달범위를 예측하기 위해서는 전파전파 모델을 이용한 예측기법이 주로 사용된다. 그러나, 전파과정에서 주변 지형지물에 의해 발생하는 전파손실은 매우 복잡한 비선형적인 특성을 가지며 수식으로는 정확한 표현이 불가능하다. 본 논문에서는 신경회로망의 함수 근사화 능력을 이용하여 전파손실 예측모델을 생성하는 방법을 제안한다. 즉, 전파손실을 송수신 안테나간의 거리, 송신안테나의 특성, 장애물 투과영향, 회절특성, 도로, 수면에 의한 영향 등과 같은 전파환경 변수들의 함수로 가정하고, 신경회로망 학습을 통하여 함수를 근사화한다. 전파환경 변수들이 신경회로망 입력으로 사용되기 위해서는 3차원 지형도와 벡터지도를 이용하여 전파의 반사, 회절, 산란 등의 물리적인 특성이 고려된 특징 추출을 통해 정량적인 수치들을 계산한다. 이와 같이 얻어진 훈련데이타를 이용한 신경회로망 학습을 통해 전파손실 모델을 완성한다. 이 모델을 이용하여 서울 도심 지역의 실제 서비스 환경에 대한 타 모델과의 비교실험결과를 통해 제안하는 모델의 우수성을 보인다.Abstract In cellular mobile communication systems, wave propagation models are used in most cases to predict cell coverage. The amount of propagation loss induced by the obstacles in the propagation path, however, is a highly non-linear function, which cannot be easily represented mathematically. In this paper, we introduce the method of producing propagation loss prediction models by function approximation using neural networks. In this method, we assume the propagation loss is a function of the relevant parameters such as the distance from the base station antenna, the specification of the transmitter antenna, obstacle profile, diffraction effect, road, and water effect. The values of these parameters are produced from the field measurement data, 3D digital terrain maps, and vector maps as its inputs by a feature extraction process, which takes into account the physical characteristics of electromagnetic waves such as reflection, diffraction and scattering. The values produced are used as the input to the neural network, which are then trained to become the propagation loss prediction model. In the experimental study, we obtain a considerable amount of improvement over COST-231 model in the prediction accuracy using this model.

• The KIPS Transactions:PartB
• /
• v.13B no.5 s.108
• /
• pp.499-506
• /
• 2006

This Paper describes a fast and stable camera pose estimation method for real-time augmented reality systems. From the feature tracking results of a marker on a single frame, we estimate the camera rotation matrix and the translation vector. For the camera pose estimation, we use the shape factorization method based on the scaled orthographic Projection model. In the scaled orthographic factorization method, all feature points of an object are assumed roughly at the same distance from the camera, which means the selected reference point and the object shape affect the accuracy of the estimation. This paper proposes a flexible and stable selection method for the reference point. Based on the proposed method, we implemented a video augmentation system that inserts virtual 3D objects into the input video frames. Experimental results showed that the proposed camera pose estimation method is fast and robust relative to the previous methods and it is applicable to various augmented reality applications.