• Database Research
• /
• v.34 no.3
• /
• pp.124-136
• /
• 2018

There have been active research activities to use neural networks to analyze OCT images and make medical decisions. One requirement for these approaches to be promising solutions is that the trained network must be generalized to new devices without a substantial loss of performance. In this paper, we use a deep convolutional neural network to distinguish AMD from normal patients. The network was trained using a data set generated from an OCT device. We observed a significant performance degradation when it was applied to a new data set obtained from a different OCT device. To overcome this performance degradation, we propose an image normalization method which performs segmentation of OCT images to identify the retina area and aligns images so that the retina region lies horizontally in the image. We experimentally evaluated the performance of the proposed method. The experiment confirmed a significant performance improvement of our approach.

• Journal of the Korea Computer Industry Society
• /
• v.10 no.5
• /
• pp.187-192
• /
• 2009

In this paper, we proposed an algorithm for personal identification of dorsal surface pattern of hand vein pattern using graph theory. Using dense ranee data images of the hand vein pattern, we used matching algorithm within the frame work of graph theory for the determination of the desired correspondence. Through preprocessing, the captured images are more sharp, clear and thinning. After thinning, the images are normalized and make graph with node and edge set. This normalized graph can make adjacent matrix. Each adjacent matrix from individual vein pattern are different. From examining the performance of individual vein patterns, we can approach performances well kind biometric technique.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.7
• /
• pp.1278-1290
• /
• 2008

This paper proposes a preprocessing method and a neural network based character recognizer to enhance the overall performance of the license plate recognition system. First, plate outlines are extracted by virtual line matching, and then the 4 vertexes are obtained by calculating intersecting points of extracted lines. By these vertexes, plate image is reconstructed as rectangle-shaped image by bilinear transform. Finally, the license plate is recognized by the neural network based classifier which had been trained using delta-bar-delta algorithm. Various license plate images were used in the experiments, and the proposed plate normalization enhanced the recognition performance up to 16 percent.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.562-564
• /
• 2002

본 논문에서는 자기상관함수의 국소적 특징을 사용하여 에지 특징을 추출한 후, 이를 이용해 유사이미지를 검색하는 방법을 제시한다. 자기상관함수의 국소적 특징을 이용하여 이미지를 검색할 경우 크기, 밝기, 색상등과 같은 이미지 요소가 서로 다를 경우에도 영향을 받지 않고 에지 특징정보를 추출해 낼 수 있다. 이는 얻어진 에지 특징을 이미지 크기와 고차 국소 자기상관함수의 변위에 의해 변하지 않도록 정규화를 하고, 동일 이미지에 대해 밝기가 조금 달라지면 검색효율이 떨어지는 점을 해결하기 위해 거리척도로서 방향여현거리(direction cosine distance)를 이용함으로써 가능하다. 이렇게 추출된 특징벡터를 자기조직화 맵에 의하여 클러스터링하고, 유사이미지 검색의 효율성을 비교해본 결과, 본 논문에서 제시한 방법을 사용하여 검색한 경우 재현율이 기존의 방법에 비해서 비교적 높은 수치를 나타냈다.

• Journal of IKEEE
• /
• v.24 no.3
• /
• pp.883-889
• /
• 2020

In this paper, we propose a method to use PV(Park's Vector) pattern for inductive motor stator fault diagnosis using CNN(Convolution Neural Network). The conventional CNN based fault diagnosis method was performed by imaging three-phase currents, but this method was troublesome to perform normalization by artificially setting the starting point and phase of current. However, when using PV pattern, the problem of normalization could be solved because the 3-phase current shows a certain circular pattern. In addition, the proposed method is proved to be superior in the accuracy of CNN by 18.18[%] compared to the previous current data image due to the autonomic normalization.

• The KIPS Transactions:PartD
• /
• v.10D no.3
• /
• pp.501-512
• /
• 2003

This paper proposes an indexing technique for fast retrieval of similar image subsequences using the multi-dimensional time warping distance. The time warping distance is a more suitable similarity measure than Lp distance in many applications where sequences may be of different lengths and/or different sampling rates. Our indexing scheme employs a disk-based suffix tree as an index structure and uses a lower-bound distance function to filter out dissimilar subsequences without false dismissals. It applies the normaliration for an easier control of relative weighting of feature dimensions and the discretization to compress the index tree. Experiments on medical and synthetic image sequences verify that the proposed method significantly outperforms the naive method and scales well in a large volume of image sequence databases.

• Journal of the Korean Geotechnical Society
• /
• v.27 no.11
• /
• pp.93-101
• /
• 2011

High quality X-ray computed microtomography (micro-CT) imaging of internal microstructures and pore space in geomaterials is often hampered by some inherent noises embedded in the images. In this paper, we introduce image calibration techniques for removing the most common noises in X-ray micro-CT, cupping (brightness difference between the periphery and central regions) and ring artifacts (consecutive concentric circles emanating from the origin). The artifacts removal sequentially applies coordinate transformation, normalization, and low-pass filtering in 2D Fourier spectrum to raw CT-images. The applicability and performance of the techniques are showcased by describing extraction of 3D pore structures from micro-CT images of porous basalt using artifacts reductions, binarization, and volume stacking. Comparisions between calibrated and raw images indicate that the artifacts removal allows us to avoid the overestimation of porosity of imaged materials, and proper calibration of the artifacts plays a crucial role in using X-ray CT for geomaterials.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.5
• /
• pp.323-330
• /
• 2020

This paper proposes the road roughness based Braking Pressure Calculation System(BPCS) for an Autonomous Vehicle Stability. The system consists of an image normalization module that processes the front image of a vehicle to fit the input of the random forest, a Random Forest based Road Roughness Classification Module that distinguish the roughness of the road on which the vehicle is travelling by using the weather information and the front image of a vehicle as an input, and a brake pressure control module that modifies a friction coefficient applied to the vehicle according to the road roughness and determines the braking strength to maintain optimal driving according to a vehicle ahead. To verify the efficiency of the BPCS experiment was conducted with a random forest model. The result of the experiment shows that the accuracy of the random forest model was about 2% higher than that of the SVM, and that 7 features should be bagged to make an accurate random forest model. Therefore, the BPCS satisfies both real-time and accuracy in situations where the vehicle needs to brake.

• Journal of IKEEE
• /
• v.26 no.2
• /
• pp.319-328
• /
• 2022

Bearing plays an important role in the operation of most machinery, Therefore, when a defect occurs in the bearing, a fatal defect throughout the machine is generated. In this reason, bearing defects should be detected early. In this paper, we describe a method using Convolutional Neural Networks (SN-CNNs) based on continuous wavelet transformations and Switchable normalization for bearing defect detection models. The accuracy of the model was measured using the Case Western Reserve University (CWRU) bearing dataset. In addition, batch normalization methods and spectrogram images are used to compare model performance. The proposed model achieved over 99% testing accuracy in CWRU dataset.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2018.07a
• /
• pp.165-166
• /
• 2018

최근 딥러닝 기술이 발전함에 따라 이를 네트워크 침입탐지 분야에 적용하려는 연구가 활발히 이루어지고 있으며 이에 따라 대용량 네트워크 데이터에 대한 처리 방법이 주목받고 있다. 본 논문에서는 네트워크 데이터를 이미지화하는 전처리 방법을 제안한다. 네트워크 데이터를 세션단위로 처리하여 손실율을 줄이면서 딥러닝 알고리즘에 바로 적용할 수 있도록 정규화된 이미지로 변환하는 방법이다. 이를 통해 딥러닝 기술을 적용한 네트워크 정보보안 분야의 연구 활성화를 기대할 수 있다.