• 대한전자공학회논문지SP
• /
• 제37권2호
• /
• pp.39-49
• /
• 2000

본 논문에서는 다중 클래스 데이터를 위한 특징 추출 방법을 최적화하는 기법을 제안한다 제안된 특징 추출 기법은 분류 오차에 기반한 방법으로 특징 공간(feature space)을 탐색하여 가우시안 최대우도 분류기 (Gaussian ML Classifier)의 분류오차(classification error)가 최소가 되도록 하는 특징벡터 집합을 구하는 방법이다 제안된 방법은 임의의 초기 특징벡터를 설정한 후 steepest descent 알고리즘을 적용하여 분류오차가 감소하는 방향으로 초기벡터를 갱신시킨다 본 논문에서는 순차탐색 및 전체탐색 두 가지의 방법을 제안하며 순차탐색은 추가로 특징벡터를 구하는 경우 이미 구해진 특징벡터를 포함하여 최소의 분류오차를 얻을 수 있는 특징벡터를 구한다 반면에 전체탐색 방법은 추가의 특징벡터를 구할 경우 새로운 초기 특징벡터 집합을 설정하여 이미 구해진 특징벡터를 포함하는 제약을 받지 않는다. 실험결과 제안된 두 가지 방법은 기존의 특징추출 방법보다 우수한 성능을 보여주고 있다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 추계종합학술대회 논문집
• /
• pp.1263-1266
• /
• 1998

In this paper, we propose an optimal feature extraction method for normally distributed multiclass data. We search the whole feature space to find a set of features that give the smallest classification error for the Gaussian ML classifier. Initially, we start with an arbitrary feature vector. Assuming that the feature vector is used for classification, we compute the classification error. Then we move the feature vector slightly and compute the classification error with this vector. Finally we update the feature vector such that the classification error decreases most rapidly. This procedure is done by taking gradient. Alternatively, the initial vector can be those found by conventional feature extraction algorithms. We propose two search methods, sequential search and global search. Experiment results show that the proposed method compares favorably with the conventional feature extraction methods.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1420-1423
• /
• 2002

The classification and recognition of two-dimensional trademark patterns independently of their position, orientation, size and scale by proposing two feature vectors has been discussed. The paper presents experimentation on two feature vectors showing size- invariance and scale-invariance respectively. Both feature vectors are equally invariant to rotation as well. The feature extraction is based on local as well as global statistics of the image. These feature vectors have appealing mathematical simplicity and are versatile. The results so far have shown the best performance of the developed system based on these unique sets of feature. The goal has been achieved by segmenting the image using connected-component (nearest neighbours) algorithm. Second part of this work considers the possibility of using back propagation neural networks (BPN) for the learning and matching tasks, by simply feeding the feature vectosr. The effectiveness of the proposed feature vectors is tested with various trademarks, not used in learning phase.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제14권3호
• /
• pp.209-215
• /
• 2014

Sequence tagging is the task of predicting frame-wise labels for a given input sequence and has important applications to diverse domains. Conventional methods such as maximum likelihood (ML) learning matches global features in empirical and model distributions, rather than local features, which directly translates into frame-wise prediction errors. Recent probabilistic sequence models such as conditional random fields (CRFs) have achieved great success in a variety of situations. In this paper, we introduce a novel discriminative CRF learning algorithm to minimize local feature mismatches. Unlike overall data fitting originating from global feature matching in ML learning, our approach reduces the total error over all frames in a sequence. We also provide an efficient gradient-based learning method via gradient forward-backward recursion, which requires the same computational complexity as ML learning. For several real-world sequence tagging problems, we empirically demonstrate that the proposed learning algorithm achieves significantly more accurate prediction performance than standard estimators.

• 전자공학회논문지B
• /
• 제29B권10호
• /
• pp.90-99
• /
• 1992

In this paper, we propose a neural network model for speech recognition. The model consists of feature extraction parts and recognition parts. The interconnection model based on ${\Delta}^2$G operator was used for frequency analysis. Two features, global feature and local feature, were extracted from this model. Recognition parts consist of global grouping stage and local grouping stage. When the input pattern was coded by slope method, the recognition rate of speakers, A and B, was 100%. When the test was performed with the data of 9 speakers, the recognition rate of 91.4% was obtained.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2012년도 제46차 하계학술발표논문집 20권2호
• /
• pp.43-46
• /
• 2012

컴퓨터 비전을 이용한 다양한 응용 분야에 있어서, 특징점을 이용한 응용 분야가 많이 이루어지고 있다. 그 중에 Global feature는 표현의 위험성과 부정확성으로 인해서 많이 사용되고 있지 않으며, Local feature를 이용한 연구가 주로 이루고 있다. 그 중에 SURF(Speeded-Up Robust Features) 알고리즘은 다수의 영상에서 같은 물리적 위치에 있는 동일한 특징점을 찾아서 매칭하는 방법으로 널리 알려진 특징점 매칭 알고리즘이다. 하지만 SURF 알고리즘을 이용하여 특징점을 매칭하여 정합 쌍을 구하였을 때 매칭되는 특징점들의 정확도가 떨어지는 단점이 있다. 본 논문에서는 특징점 매칭 알고리즘인 SURF를 사용하여 대응되는 특징점들을 들로네 삼각형의 기하학적 특징을 이용하여 정확도가 높은 특징점을 분류하여 SURF 알고리즘의 매칭되는 대응점들의 정확도를 높이는 방법을 제안한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권9호
• /
• pp.4412-4428
• /
• 2018

The purpose of this work is to solve the problem of representing an entire video using Convolutional Neural Network (CNN) features for human action recognition. Recently, due to insufficient GPU memory, it has been difficult to take the whole video as the input of the CNN for end-to-end learning. A typical method is to use sampled video frames as inputs and corresponding labels as supervision. One major issue of this popular approach is that the local samples may not contain the information indicated by the global labels and sufficient motion information. To address this issue, we propose a binary hashing method to enhance the local feature extractors. First, we extract the local features and aggregate them into global features using maximum/minimum pooling. Second, we use the binary hashing method to capture the motion features. Finally, we concatenate the hashing features with global features using different normalization methods to train the classifier. Experimental results on the JHMDB and MPII-Cooking datasets show that, for these new local features, binary hashing mapping on the sparsely sampled features led to significant performance improvements.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권1호
• /
• pp.105-125
• /
• 2024

Gliomas are the most common malignant brain tumor and cause the most deaths. Manual brain tumor segmentation is expensive, time-consuming, error-prone, and dependent on the radiologist's expertise and experience. Manual brain tumor segmentation outcomes by different radiologists for the same patient may differ. Thus, more robust, and dependable methods are needed. Medical imaging researchers produced numerous semi-automatic and fully automatic brain tumor segmentation algorithms using ML pipelines and accurate (handcrafted feature-based, etc.) or data-driven strategies. Current methods use CNN or handmade features such symmetry analysis, alignment-based features analysis, or textural qualities. CNN approaches provide unsupervised features, while manual features model domain knowledge. Cascaded algorithms may outperform feature-based or data-driven like CNN methods. A revolutionary cascaded strategy is presented that intelligently supplies CNN with past information from handmade feature-based ML algorithms. Each patient receives manual ground truth and four MRI modalities (T1, T1c, T2, and FLAIR). Handcrafted characteristics and deep learning are used to segment brain tumors in a Global Convolutional Neural Network (GCNN). The proposed GCNN architecture with two parallel CNNs, CSPathways CNN (CSPCNN) and MRI Pathways CNN (MRIPCNN), segmented BraTS brain tumors with high accuracy. The proposed model achieved a Dice score of 87% higher than the state of the art. This research could improve brain tumor segmentation, helping clinicians diagnose and treat patients.

• 한국CDE학회논문집
• /
• 제5권1호
• /
• pp.12-22
• /
• 2000

Network and Internet technology opens up another domain for building future CAD/CAM environment. The environment will be global, network-centric, and spatially distributed. In this paper, we present an approach for network-centric feature-based modeling in a distributed design environment. The presented approach combines the current feature-based modeling technique with distributed computing and communication technology for supporting product modeling and collaborative design activities over the network. The approach is implemented in a client/server architecture, in which Web-enabled feature modeling clients, neutral feature model server, and other applications communicate with one another via a standard communication protocol. The paper discusses how the neutral feature model supports multiple views and maintains naming consistency between geometric entities of the server and clients. Moreover, it explains how to minimize the network delay between the server and client according to incremental feature modeling operations.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 학술회의 논문집 정보 및 제어부문 A
• /
• pp.11-14
• /
• 2003

In this paper, we use Fuzzy Logic and Potential field method for optimal path planning of an autonomous mobile robot and apply to navigation for real-time mobile robot in 2D dynamic environment. For safe navigation of the robot, we use both Global and Local path planning. Global path planning is computed off-line using sell-decomposition and Dijkstra algorithm and Local path planning is computed on-line with sensor information using potential field method and Fuzzy Logic. We can get gravitation between two feature points and repulsive force between obstacle and robot through potential field. It is described as a summation of the result of repulsive force between obstacle and robot which is considered as an input through Fuzzy Logic and gravitation to a feature point. With this force, the robot fan get to desired target point safely and fast avoiding obstacles. We Implemented the proposed algorithm with Pioneer-DXE robot in this paper.