• 대한의용생체공학회:의공학회지
• /
• 제43권6호
• /
• pp.441-447
• /
• 2022

In this paper, we propose a pill identification model using engraved text feature and image feature such as shape and color, and compare it with an identification model that does not use engraved text feature to verify the possibility of improving identification performance by improving recognition rate of the engraved text. The data consisted of 100 classes and used 10 images per class. The engraved text feature was acquired through Keras OCR based on deep learning and 1D CNN, and the image feature was acquired through 2D CNN. According to the identification results, the accuracy of the text recognition model was 90%. The accuracy of the comparative model and the proposed model was 91.9% and 97.6%. The accuracy, precision, recall, and F1-score of the proposed model were better than those of the comparative model in terms of statistical significance. As a result, we confirmed that the expansion of the range of feature improved the performance of the identification model.

• Industrial Engineering and Management Systems
• /
• 제16권1호
• /
• pp.64-79
• /
• 2017

Identifying disease genes from human genome is a critical task in biomedical research. Important biological features to distinguish the disease genes from the non-disease genes have been mainly selected based on traditional feature selection approaches. However, the traditional feature selection approaches unnecessarily consider many unimportant biological features. As a result, although some of the existing classification techniques have been applied to disease gene identification, the prediction performance was not satisfactory. A small set of the most important biological features can enhance the accuracy of disease gene identification, as well as provide potentially useful knowledge for biologists or clinicians, who can further investigate the selected biological features as well as the potential disease genes. In this paper, we propose a new stepwise random forests (SRF) approach for biological feature selection and disease gene identification. The SRF approach consists of two stages. In the first stage, only important biological features are iteratively selected in a forward selection manner based on one-dimensional random forest regression, where the updated residual vector is considered as the current response vector. We can then determine a small set of important biological features. In the second stage, random forests classification with regard to the selected biological features is applied to identify disease genes. Our extensive experiments show that the proposed SRF approach outperforms the existing feature selection and classification techniques in terms of biological feature selection and disease gene identification.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 Proceedings of the Korea Automatic Control Conference, 11th (KACC); Pohang, Korea; 24-26 Oct. 1996
• /
• pp.145-148
• /
• 1996

In moving object tracking based on the visual sensory feedback, a prerequisite is to determine which feature or which object is to be tracked and then the feature or the object identification precedes the tracking. In this paper, we focus on the object identification not image feature identification. The target identification is realized by finding out corresponding line segments to the hypothesized model segments of the target. The key idea is the combination of the Mahalanobis distance with the geometrica relationship between model segments and extracted line segments. We demonstrate the robustness and feasibility of the proposed target identification algorithm by a moving vehicle identification and tracking in the video traffic surveillance system over images of a road scene.

• 한국정보과학회논문지:데이타베이스
• /
• 제41권4호
• /
• pp.199-208
• /
• 2014

본 논문에서는 스캔된 만화의 콘텐츠 특성을 고려한 식별 및 특징 검색 시스템을 제안하였다. 스캔 만화의 특징점을 생성하기 위해서 계층적 대칭 핑거프린팅 방법을 활용하였다. 제안하는 핑거프린트 식별 및 검색 시스템은 웹하드와 같은 온라인 서비스 제공자들이 대량의 스캔만화에 대하여 즉각적인 식별 결과를 얻을 수 있도록 설계되었다. 실험에서는 회전, 이동 등의 이미지 변형에 대해서 핑거프린트의 식별 강인성에 대하여 분석하였다. 또한 특징점 데이터베이스에서의 빠른 매칭을 위한 데이터베이스 구조를 제안하였고, 전역 검색 및 최대중요특징 검색과 같은 기존의 다른 검색방법과 성능을 비교하였다.

• 대한전자공학회논문지SP
• /
• 제48권4호
• /
• pp.76-85
• /
• 2011

본 논문에서는 Gabor 특징과 웨이브렛 영역의 BDIP와 BVLC 특징을 이용한 질감 특징 기반 언어 인식 방법을 제안한다. 제안된 방법에서는 먼저 시험 영상에 Gabor 변환과 웨이브렛 변환을 적용한다. 웨이브렛 영역의 상세 대역에는 Donoho의 연역치화를 적용하여 잡음을 제거한다. 이어서 Gabor 영상에는 크기 연산자를 적용하고 웨이브렛 부대역에는 BDIP와 BVLC 연산자를 적용한다. 그런 다음 Gabor 크기 영상과 BDIP, BVLC 부대역에 대하여 통계치를 계산하여 그 결과들을 벡터화하고 융합하여 특징 벡터로 사용한다. 분류 단계에서는 얼굴 인식에 주로 사용되는 WPCA를 분류기로 하여 시험 특징 벡터와 가장 유사한 학습 특징 벡터를 찾는다. 실험 결과 제안된 방법은 실험 문서 영상 DB에 대하여 비교적 낮은 특징 벡터 차원으로 매우 우수한 언어 인식 성능을 보여준다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.699-702
• /
• 2009

This paper proposes a gender identification using shoeprint images. It is difficult for the proposed method to identify an individual if shoeprint images for identification leaked out. Because the proposed method identifies gender without the faces, the type of dress and the hair types images. Therefore we can use safely the proposed method in public place. In addition, a sensor mat which we developed is reasonable to use mechanical switches arranged in a matrix pattern without pressure switches. We had shoeprint images with the sensor mat. We measure feature parameters from shoeprint images. The feature parameters are length, width and area of shoeprint. Utilizing the feature parameters, we identified gender. In order to verify the gender identification rate of the proposed method, we set up the sensor mat at an entrance of buildings and took shoeprint images of 100 men and 100 women. As a result, we achieved about 86 percent of the gender identification rate.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.2213-2216
• /
• 2003

In this paper, we propose the face identification method which is robust for lighting based on the feature points method. First of all, the proposed method extracts an edge of facial feature. Then, by the hough transform, it determines ellipse parameters of each facial feature from the extracted edge. Finally, proposed method performs the face identification by using parameters. Even if face image is taken under various lighting condition, it is easy to extract the facial feature edge. Moreover, it is possible to extract a subject even if the object has not appeared enough because this method extracts approximately the parameters by the hough transformation. Therefore, proposed method is robust for the lighting condition compared with conventional method. In order to show the effectiveness of the proposed method, computer simulations are done by using the real images.

• 말소리와 음성과학
• /
• 제1권1호
• /
• pp.69-73
• /
• 2009

This paper proposes an efficient global covariance-based principal component analysis (GCPCA) for speaker identification. Principal component analysis (PCA) is a feature extraction method which reduces the dimension of the feature vectors and the correlation among the feature vectors by projecting the original feature space into a small subspace through a transformation. However, it requires a larger amount of training data when performing PCA to find the eigenvalue and eigenvector matrix using the full covariance matrix by each speaker. The proposed method first calculates the global covariance matrix using training data of all speakers. It then finds the eigenvalue matrix and the corresponding eigenvector matrix from the global covariance matrix. Compared to conventional PCA and Gaussian mixture model (GMM) methods, the proposed method shows better performance while requiring less storage space and complexity in speaker identification.

• 한국멀티미디어학회논문지
• /
• 제17권3호
• /
• pp.277-286
• /
• 2014

본 논문에서는 Gabor 특징과 MDLC 특징, 그리고 co-occurrence 특징의 융합에 의한 질감 특징 기반언어 인식 방법을 제안한다. 제안된 방법에서는 먼저 시험 영상에 Gabor 변환에 이은 크기 연산자를 적용하여 Gabor 크기 영상을 얻고 그 통계치를 계산하여 결과를 벡터화한다. 이어서 MDLC 연산자를 이용하여 MDLC 영상을 얻고 역시 그 통계치를 계산하여 벡터화한다. 다음으로 시험 영상으로부터 GLCM을 계산하고 이를 이용하여 co-occurrence 특징을 계산한 다음 벡터화한다. 이들 Gabor, MDLC, co-occurrence 특징에 의한 벡터들은 벡터 융합에 의하여 특징 벡터로 사용된다. 분류 단계에서는 얼굴 인식에 주로 사용되는 WPCA를 분류기로 하여 시험 특징 벡터와 가장 유사한 학습 특징 벡터를 찾는다. 제안된 방법의 성능은 15개국 언어의 문서를 스캔하여 얻은 시험 문서 영상 DB에 대한 평균 인식률을 조사하여 알아본다. 실험 결과 제안된 방법은 시험 DB에 대하여 비교적 낮은 특징 벡터 차원으로 매우 우수한 언어 인식 성능을 보여준다.

• Journal of Information Processing Systems
• /
• 제16권2호
• /
• pp.435-446
• /
• 2020

With the ability to learn rules from training data, the machine learning model can classify unknown objects. At the same time, the dimension of hyperspectral data is usually large, which may cause an over-fitting problem. In this research, an identification methodology of tea diseases was proposed based on spectral reflectance and machine learning, including the feature selector based on the decision tree and the tea disease recognizer based on random forest. The proposed identification methodology was evaluated through experiments. The experimental results showed that the recall rate and the F1 score were significantly improved by the proposed methodology in the identification accuracy of tea disease, with average values of 15%, 7%, and 11%, respectively. Therefore, the proposed identification methodology could make relatively better feature selection and learn from high dimensional data so as to achieve the non-destructive and efficient identification of different tea diseases. This research provides a new idea for the feature selection of high dimensional data and the non-destructive identification of crop diseases.