• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권6호
• /
• pp.1833-1848
• /
• 2022

Named entity recognition (NER) is an important basic task in the field of Natural Language Processing (NLP). Recently deep learning approaches by extracting word segmentation or character features have been proved to be effective for Chinese Named Entity Recognition (CNER). However, since this method of extracting features only focuses on extracting some of the features, it lacks textual information mining from multiple perspectives and dimensions, resulting in the model not being able to fully capture semantic features. To tackle this problem, we propose a novel Multi-view Semantic Feature Fusion Model (MSFM). The proposed model mainly consists of two core components, that is, Multi-view Semantic Feature Fusion Embedding Module (MFEM) and Multi-head Self-Attention Mechanism Module (MSAM). Specifically, the MFEM extracts character features, word boundary features, radical features, and pinyin features of Chinese characters. The acquired font shape, font sound, and font meaning features are fused to enhance the semantic information of Chinese characters with different granularities. Moreover, the MSAM is used to capture the dependencies between characters in a multi-dimensional subspace to better understand the semantic features of the context. Extensive experimental results on four benchmark datasets show that our method improves the overall performance of the CNER model.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2008년도 춘계종합학술대회 A
• /
• pp.151-154
• /
• 2008

본 논문은 현 검색시스템의 단순한 키워드 입력 방식에서 발생하는 오차를 줄이기 위해 이미지의 그레이스케일 히스토그램과 에지정보를 이용하는 검색 시스템 구현을 하였다. 검색알고리즘은 질의 이미지의 특징을 추출하는 단계, 이미지 정제 및 에지정보 추출단계, 추출된 특징을 분석하는 단계, 분석된 특징들로부터 필요한 정보를 확보하는 단계, 확보된 정보를 데이터베이스로부터 검색하는 단계, 검색된 데이터베이스에서 이미지를 비교 추출단계로 이루어진다. 제안한 검색시스템은 빠른 검색과 고 정확도를 목적으로 실현되며 시뮬레이션을 통해 이를 검증하고자 하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 추계종합학술대회 논문집(4)
• /
• pp.65-68
• /
• 2000

This paper describes an approach for extracting invariant features using a view-based representation and recognizing an object with a high speed search method in FLIR. In this paper, we use a reformulated eigenspace technique based on robust estimation for extracting features which are robust for outlier such as noise and clutter. After extracting feature, we recognize an object using a partial distance search method for calculating Euclidean distance. The experimental results show that the proposed method achieves the improvement of recognition rate compared with standard PCA.

• 한국통신학회논문지
• /
• 제19권2호
• /
• pp.373-380
• /
• 1994

본 논문에서는 음성신호를 일정한 크기로 적응시켜 최적의 특징을 추출할 수 있는 방법을 제안하였다. 음성신호의 특징을 추출하기 위하여 고속선형예측 알고리즘인 FRLS 적용할 때 음성신호를 일정한 크기로 분할한 후 각 프레임 마다 제안한 균등사기상관함수를 가지고 최적특징을 추출하였다.

• 대한전자공학회논문지
• /
• 제25권5호
• /
• pp.559-565
• /
• 1988

A new valley operator is presented for extracting sketch features which contain valleys and edges subject to local intensities. It is a very simple operator using the local probablities in a 3x3 local window. Experimental results show its excellent performance over the existing valley or edge operators.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2006년도 Proceedings of ISRS 2006 PORSEC Volume I
• /
• pp.181-183
• /
• 2006

Digital topographic map in Korea contains layers of spatial and attribute data for 8 land features such as railroads, watercourses, roads, buildings and etc. Some of the layers such as building and forest don't include any information about height, which can be just prepared by interpretation of remote sensed data or field survey. LiDAR(Light Detection And Ranging) data using active pulse and digital camera provides data about height and form of land features. LiDAR data can be used not only to extract the outline of land features but also to estimate the height. This study presents technical availability for extraction and estimation of land feature's outline and height using LiDAR data which composes of natural and artificial land features, and digital aerial photograph which was taken simultaneously with the LiDAR. The estimated location, outline and height of land features were compared with the field survey data, and we could find that LiDAR data and digital aerial photograph can be a useful source for estimating the height of land features as well as extracting the outline.

• 한국농업기계학회:학술대회논문집
• /
• 한국농업기계학회 1996년도 International Conference on Agricultural Machinery Engineering Proceedings
• /
• pp.773-779
• /
• 1996

Dried oak mushroom have complex and various visual features. Grading and sorting of dried oak mushrooms has been done by the human expert. Though actions involved in human grading looked simple, a decision making underneath the simple action comes from the result of the complex neural processing of the visual image. Through processing details involved in human visual recognition has not been fully investigated yet, it might say human can recognize objects via one of three ways such as extracting specific features or just image itself without extracting those features or in a combined manner. In most cases, extracting some special quantitative features from the camera image requires complex algorithms and processing of the gray level image requires the heavy computing load. This fact can be worse especially in dealing with nonuniform, irregular and fuzzy shaped agricultural products, resulting in poor performance because of the sensitiveness to the crisp criteria or specific ules set up by algorithms. Also restriction of the real time processing often forces to use binary segmentation but in that case some important information of the object can be lost. In this paper, the neuro net based real time recognition algorithm was proposed without extracting any visual feature but using only the directly captured raw gray images. Specially formated adaptable size of grids was proposed for the network input. The compensation of illumination was also done to accomodate the variable lighting environment. The proposed grading scheme showed very successful results.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(3)
• /
• pp.77-80
• /
• 2001

In general, the informations of the inner image that user interested in are limited to a special domain. In this paper, as using Wavelet Transform for dividing image into high frequency and low frequency, We can separate foreground including many data. After calculating object boundary of separated part, We extract special features using Color Coherence Vector. According to results of this experiment, the method of comparing data extracting foreground features is more effective than comparing data extracting features of entire image when we extract the image user interested in.

• 한국멀티미디어학회:학술대회논문집
• /
• 한국멀티미디어학회 2000년도 춘계학술발표논문집
• /
• pp.328-331
• /
• 2000

In this paper , we propose a texture feature extraction method of reduce the massive computational time on extracting texture, features of large sized medical such as MRI, CT-scan , and an index structure, called GLTFT, to speed up the retrieval performance. For these, the original image is transformed into a compressed image by Wavelet transform , and textural features such as contrast, energy, entropy, and homogeneity of the compressed image is extracted by using GLCM(Gray Level Co-occurrence Metrix) . The proposed index structure is organized by using the textural features. The processing in compressed domain can give the solution of storage space and the reduction of computational time of feature extracting . And , by GLTFT index structure, image retrieval performance can be expected to be improved by reducing the retrieval range . Our experiment on 270 MRIs as image database shows that shows that such expectation can be got.

• 한국정보처리학회논문지
• /
• 제3권6호
• /
• pp.1607-1615
• /
• 1996

본 논문에서는 계산이 간단하면서 효과적으로 스케치 특징을 추출할수 있는 새로운 연산자 DBAH(differ-ence between arithmetic mean and harmonic mean) 를 제안하고,임계치를 자동으로 결정하는 퍼지 임계치 추론기를 제안한다. 제안한 연산자는 국부 밝기를 고려하며,매우 어두운 영역에서의 밝기변화에서는 낮은 출력을 나타내는 장점을 가진다. 또한 제안된 퍼지 임계치 추론기는 조작자의 개입 없이도 스케치 특징점을 매우 잘 검출한다.