• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.1-3
• /
• 2022

최근 기존의 영상 압축 파이프라인 대신 신경망의 종단 간 학습을 통해 압축을 수행하는 알고리즘의 연구가 활발히 진행되고 있다. 본 논문은 종단 간 학습 기반 공간적 스케일러블 압축 기술을 제안한다. 보다 구체적으로 본 논문은 신경망의 각 계층에서 하위 계층의 학습된 특징 (feature)을 융합하여 상위 계층으로 전달하는 다중 스케일 특징 융합 (multi-scale feature fusion) 모듈을 도입해 상위 계층이 더욱 풍부한 특징 정보를 학습하고 계층 사이의 특징 중복성을 더욱 잘 제거할 수 있도록 한다. 기존 방법 대비 향상 계층(enhancement layer)에서 1.37%의 BD-rate가 향상된 결과를 볼 수 있다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.3 no.3
• /
• pp.27-35
• /
• 2002

A knowledge-based image retrieval technique is image searching methods using some features from the queried image. The materials in this study are automobile head lamps. The input data is composed of characters and images which have various pattern. The numbers, special symbols, and general letters are under the category of the character. The image informations are made up of the distribution of pixel data, statistical analysis, and state of pattern which are useful for the knowledge data. In this paper, we implemented a retrieval system for the scientific crime detection at traffic accident using the proposed multi knowledge-based image retrieval technique. The values for the multi knowledge-based image features were extracted from color and gray scale each. With this 22 features, we improved the retrieval efficiency about the color information and pattern information. Visual basic, crystal report and MS access DB were used for this application. We anticipate the efficient scientific detection for the traffic accident and the tracking of suspicious vehicle.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.6
• /
• pp.43-49
• /
• 2004

The classification time required by conventional multi-class SVMs(Support Vector Machines) greatly increases as the number of pattern classes increases. This is due to the fact that the needed set of binary class SVMs gets quite large. In this paper, we propose a method to reduce the number of classes by using nearest neighbor rule (NNR) in the principle component analysis and linear discriminant analysis (PCA+LDA) feature subspace. The proposed method reduces the number of face classes by selecting a few classes closest to the test data projected in the PCA+LDA feature subspace. Results of experiment show that our proposed method has a lower error rate than nearest neighbor classification (NNC) method. Though our error rate is comparable to the conventional multi-class SVMs, the classification process of our method is much faster.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.11
• /
• pp.95-101
• /
• 1999

In this paper, we propose an adaptive method of tracking multiple moving objects using contour and features in surrounding conditions. We use an adaptive background model for robust processing in surrounding conditions. Object segmentation model detects pixels thresholded from local difference image between background and current image and extracts connected regions. Data association problem is solved by using feature extraction and object recognition model in searching window. We use Kalman filters for real-time tracking. The results of simulation show that the proposed method is good for tracking multiple moving objects in highway image sequences.

• Journal of Korea Multimedia Society
• /
• v.16 no.8
• /
• pp.914-926
• /
• 2013

Among the recent security systems, security system with fingerprint recognition gets many people's interests through the strengths such as exclusiveness, convenience, etc, in comparison with other security systems. The most important matters for fingerprint recognition system are reliability of matching between the fingerprint in database and user's fingerprint and rapid process of image processing algorithms used for fingerprint recognition. The existing fingerprint recognition system reduces the processing time by removing some processes in the feature extraction algorithms but has weakness of a reliability. This paper realizes the fingerprint recognition algorithm using MAMS(Multi-Access Memory System) for both the rapid processing time and the reliability in feature extraction and matching accuracy. Reliability of this process is verified by the correlation between serial processor's results and MAMS-PP64's results. The performance of the method using MAMS-PP64 is 1.56 times faster than compared serial processor.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.2
• /
• pp.160-166
• /
• 2020

As mobile-based hardware technology develops, many kinds of applications are also being developed. In addition, there is an increasing demand to automatically check that the interface of these applications works correctly. In this paper, we describe a method for accurately detecting faulty images from applications by comparing major characteristics from input color images. For this purpose, our method first extracts major characteristics of the input image, then calculates the differences in the extracted major features, and decides if the test image is a normal image or a faulty image dissimilar to the reference image. Experiment results show that the suggested approach robustly determines similar and dissimilar images by comparing major characteristics from input color images. The suggested method is expected to be useful in many real application areas related to computer vision, like video indexing, object detection and tracking, image surveillance, and so on.

• The Journal of the Korea Contents Association
• /
• v.7 no.1
• /
• pp.48-57
• /
• 2007

Eye localization is necessary for face recognition and related application areas. Most of eye localization algorithms reported thus far still need to be improved about precision and computational time for successful applications. In this paper, we propose an improved eye localization method based on multi-scale Gator feature vector models. The proposed method first tries to locate eyes in the downscaled face image by utilizing Gabor Jet similarity between Gabor feature vector at an initial eye coordinates and the eye model bunch of the corresponding scale. The proposed method finally locates eyes in the original input face image after it processes in the same way recursively in each scaled face image by using the eye coordinates localized in the downscaled image as initial eye coordinates. Experiments verify that our proposed method improves the precision rate without causing much computational overhead compared with other eye localization methods reported in the previous researches.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.4
• /
• pp.598-603
• /
• 2003

This paper presents a sub-band feature extraction approach in which the feature extraction method in the individual frequency sub-bands is determined in terms of speech recognition accuracy. As in the multi-band paradigm, features are extracted independently in frequency sub-regions of the speech signal. Since the spectral shape is well structured in the low frequency region, the all pole model is effective for feature extraction. But, in the high frequency region, the nonparametric transform, discrete cosine transform is effective for the extraction of cepstrum. Using the sub-band specific feature extraction method, the linguistic information in the individual frequency sub-bands can be extracted effectively for automatic speech recognition. The validity of the proposed method is shown by comparing the results of speech recognition experiments for our method with those obtained using a full-band feature extraction method.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2002.05a
• /
• pp.215-220
• /
• 2002

본 연구에서는 다중채널 뇌파에서 특징 파라미터로 선형 예측기 계수(Linear predictor coefficients)를 추출하고, 패턴인식기로는 신경회로망을 이용한 쾌적성 분류 알고리즘을 개발하여 다중 템플릿 방법으로 쾌적성 분류 실험을 하고자 하였다. 뇌파 데이터는 대학생 10명으로부터 쾌적한 환경과 불쾌적한 환경에서의 데이터를 수집하였으며, 전극 위치는 Fpl, Fp2, F3, F4, T3, T4, P3, P4, O1, O2를 사용하였다. 수집된 뇌파는 전처리를 거친 후 특징 파라미터를 추출하고 패턴 분류기로 사용된 신경회로망의 입력으로 사용하였다. 쾌적성 분류 방법은 다중템플릿 방법으로 여러 명의 피검자를 각각 학습시켜 이로부터 생성되는 신경회로망의 가중치들을 템플릿에 저장한다. 그리고 테스트를 할 때에는 먼저 처음의 안정 상태의 뇌파를 이용하여 템플릿 검색을 하고 가장 가까운 템플릿을 선택한다. 그리고 선택된 템플릿을 이용하여 다른 감정에 대한 쾌적성 분류 실험을 하게 된다. 쾌적성 분류 실험 결과 평균 인식률이 약 75%의 성능을 나타내었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.04a
• /
• pp.187-190
• /
• 2002

본 논문은 다중 특징을 포함한 대표 키 프레임을 추출을 통해 장면 전환 검출 시 발생할 수 있는 검출 오류를 자동으로 인식하고 수정함으로써 빠르고 신뢰성 있는 장면 분할을 수행하는 새로운 기법을 제안한다. 이를 위해 개선된 고속 장면 전환 검출 기법에 의해 샷을 분할 하고 분할 된 샷으로부터 대표 키 프레임과 그것에 포함된 후보 키 프레임들의 다중 정보를 포함시킴으로써 샷의 전반에 대한 정보를 보다 잘 표현할 수 있도록 한다. 그리고 다중정보를 포함한 대표 키 프레임의 비교를 통해 샷 검출 오류를 자동으로 인식하여 적절히 수정할 수 있는 기법을 제안하며 실세계 동영상 데이터를 사용한 실험을 통해서 제안하는 기법에 의해 효율적으로 샷이 분할 될 수 있음을 보인다.