• 제어로봇시스템학회:학술대회논문집
• /
• 1994.10a
• /
• pp.302-306
• /
• 1994

This paper describes how to recognize hand written Hangeul character using the stroke order of the elementary segment. The recognition system is constructed of parts : character input part, segment disassembling part, character element extraction part and character recognition part. The character input part reads the character and performs thinning algorithm. In the segment disassembling part, the input character is disassembled into elementary segments using the direction codes and the feature parameters. In the character element extraction part, we extract the character element using the stroke order and the knowledge rule. Finally, we able to recognize the hand written Hangeul characters by assembling the character elements, in the character recognition part.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.919-920
• /
• 2008

In this paper, we present early processing techniques for visual inspection of metallic parts. Since metallic surfaces give rise to specular reflections, it is difficult to extract object boundaries using elementary segmentation techniques such as edge detection or binary thresholding. In this paper, we present two techniques for finding object boundaries on micro bit images. First, we explain a technique for detecting blade boundaries using a directional correlation mask. Second, a line and angle extraction technique based on Harris corner detector and Hough transform is described. These techniques have been effective for detecting blade boundaries, and a number of experimental results are presented using real images.

• Journal of Electrical Engineering and Technology
• /
• v.11 no.6
• /
• pp.1857-1862
• /
• 2016

Human activity recognition using depth information is an emerging and challenging technology in computer vision due to its considerable attention by many practical applications such as smart home/office system, personal health care and 3D video games. This paper presents a novel framework of 3D human body detection, tracking and recognition from depth video sequences using spatiotemporal features and modified HMM. To detect human silhouette, raw depth data is examined to extract human silhouette by considering spatial continuity and constraints of human motion information. While, frame differentiation is used to track human movements. Features extraction mechanism consists of spatial depth shape features and temporal joints features are used to improve classification performance. Both of these features are fused together to recognize different activities using the modified hidden Markov model (M-HMM). The proposed approach is evaluated on two challenging depth video datasets. Moreover, our system has significant abilities to handle subject's body parts rotation and body parts missing which provide major contributions in human activity recognition.

• Journal of Korea Society of Industrial Information Systems
• /
• v.6 no.2
• /
• pp.16-22
• /
• 2001

In this paper we mose a new scheme for recognition of deformed letters by extracting feature vectors based on Gibbs distributions which are well suited for representing the spatial continuity. The extracted feature vectors are comprised of 2-D conditional moments which are invariant under translation, rotation, and scale of an image. The Algorithm for pattern recognition of deformed letters contains two parts: the extraction of feature vector and the recognition process. (i) We extract feature vector which consists of an improved 2-D conditional moments on the basis of estimated conditional Gibbs distribution for an image. (ii) In the recognition phase, the minimization of the discrimination cost function for a deformed letters determines the corresponding template pattern. In order to evaluate the performance of the proposed scheme, recognition experiments with a generated document was conducted. on Workstation. Experiment results reveal that the proposed scheme has high recognition rate over 96%.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.2 s.40
• /
• pp.159-167
• /
• 2006

Lipreading has been suggested as one of the methods to improve the performance of speech recognition in noisy environment. However, existing methods are developed and implemented only in software. This paper suggests a hardware design for real-time lipreading. For real-time processing and feasible implementation, we decompose the lipreading system into three parts; image acquisition module, feature vector extraction module, and recognition module. Image acquisition module capture input image by using CMOS image sensor. The feature vector extraction module extracts feature vector from the input image by using parallel block matching algorithm. The parallel block matching algorithm is coded and simulated for FPGA circuit. Recognition module uses HMM based recognition algorithm. The recognition algorithm is coded and simulated by using DSP chip. The simulation results show that a real-time lipreading system can be implemented in hardware.

• Journal of KIISE:Software and Applications
• /
• v.35 no.12
• /
• pp.789-799
• /
• 2008

For the communication between human and computer in an interactive computing environment, the gesture recognition has been studied vigorously. The algorithms which use the 2D features for the feature extraction and the feature comparison are faster, but there are some environmental limitations for the accurate recognition. The algorithms which use the 2.5D features provide higher accuracy than 2D features, but these are influenced by rotation of objects. And the algorithms which use the 3D features are slow for the recognition, because these algorithms need the 3d object reconstruction as the preprocessing for the feature extraction. In this paper, we propose a method to extract the 3D features combined with the 3D object reconstruction in real-time. This method generates three kinds of 3D projection maps using the modified GPU-based visual hull generation algorithm. This process only executes data generation parts only for the gesture recognition and calculates the Hu-moment which is corresponding to each projection map. In the section of experimental results, we compare the computational time of the proposed method with the previous methods. And the result shows that the proposed method can apply to real time gesture recognition environment.

• Journal of KIISE:Software and Applications
• /
• v.31 no.10
• /
• pp.1412-1417
• /
• 2004

Sparse coding or independent component analysis (ICA) which is a holistic representation, was successfully applied to elucidate early auditor${\gamma}$ processing and to the task of sound classification. In contrast, parts-based representation is an alternative way o) understanding object recognition in brain. In this thesis we employ the non-negative matrix factorization (NMF) which learns parts-based representation in the task of sound classification. Methods of feature extraction from the spectro-temporal sounds using the NMF in the absence or presence of noise, are explained. Experimental results show that NMF-based features improve the performance of sound classification over ICA-based features.

• The Journal of Korean Institute of Information Technology
• /
• v.16 no.11
• /
• pp.105-113
• /
• 2018

In this paper, we proposed a method for effective classification of eye, nose, and mouth of human face. Most recent image classification uses Convolutional Neural Network(CNN). However, the features extracted by CNN are not sufficient and the classification effect is not too high. We proposed a new algorithm to improve the classification effect. The proposed method can be roughly divided into three parts. First, the Haar feature extraction algorithm is used to construct the eye, nose, and mouth dataset of face. The second, the model extracts CNN features of image using AlexNet. Finally, Haar-CNN features are extracted by performing convolution after Haar feature extraction. After that, CNN features and Haar-CNN features are fused and classify images using softmax. Recognition rate using mixed features could be increased about 4% than CNN feature. Experiments have demonstrated the performance of the proposed algorithm.

• The Journal of the Korea Contents Association
• /
• v.7 no.6
• /
• pp.52-59
• /
• 2007

This paper presents a greedy method for user-steered mesh segmentation, which is based on the merging priority metric defined for representing the geometric properties of meaningful parts. The priority metric is a weighted function composed of five geometric parameters: distribution of Gaussian map, boundary path concavity, boundary path length, cardinality, and segmentation resolution. This scheme can be extended without any modification only by defining more geometric parameters and adding them. Our experimental results show that the shapes of segmented parts can be controlled by setting up the weight values of geometric parameters.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.1
• /
• pp.1-15
• /
• 2022

The MobileNetV3 is specially designed for mobile devices with limited memory and computing power. To reduce the network parameters and improve the network inference speed, a new lightweight network is proposed based on MobileNetV3. Firstly, to reduce the computation of residual blocks, a partial residual structure is designed by dividing the input feature maps into two parts. The designed partial residual structure is used to replace the residual block in MobileNetV3. Secondly, a dual-path feature extraction structure is designed to further reduce the computation of MobileNetV3. Different convolution kernel sizes are used in the two paths to extract feature maps with different sizes. Besides, a transition layer is also designed for fusing features to reduce the influence of the new structure on accuracy. The CIFAR-100 dataset and Image Net dataset are used to test the performance of the proposed partial residual structure. The ResNet based on the proposed partial residual structure has smaller parameters and FLOPs than the original ResNet. The performance of improved MobileNetV3 is tested on CIFAR-10, CIFAR-100 and ImageNet image classification task dataset. Comparing MobileNetV3, GhostNet and MobileNetV2, the improved MobileNetV3 has smaller parameters and FLOPs. Besides, the improved MobileNetV3 is also tested on CPU and Raspberry Pi. It is faster than other networks