• Journal of Information Processing Systems
• /
• v.16 no.2
• /
• pp.343-359
• /
• 2020

In this paper, a new algorithm is proposed for three-dimensional (3D) shape recognition using local features of model views and its sparse representation. The algorithm starts with the normalization of 3D models and the extraction of 2D views from uniformly distributed viewpoints. Consequently, the 2D views are stacked over each other to from view cubes. The algorithm employs the descriptors of 3D local features in the view cubes after applying Gabor filters in various directions as the initial features for 3D shape recognition. In the training stage, we store some 3D local features to build the prototype dictionary of local features. To extract an intermediate feature vector, we measure the similarity between the local descriptors of a shape model and the local features of the prototype dictionary. We represent the intermediate feature vectors of 3D models in the sparse domain to obtain the final descriptors of the models. Finally, support vector machine classifiers are used to recognize the 3D models. Experimental results using the Princeton Shape Benchmark database showed the average recognition rate of 89.7% using 20 views. We compared the proposed approach with state-of-the-art approaches and the results showed the effectiveness of the proposed algorithm.

• Journal of Information Processing Systems
• /
• v.12 no.3
• /
• pp.468-488
• /
• 2016

The paper proposes a novel framework for 3D face verification using dimensionality reduction based on highly distinctive local features in the presence of illumination and expression variations. The histograms of efficient local descriptors are used to represent distinctively the facial images. For this purpose, different local descriptors are evaluated, Local Binary Patterns (LBP), Three-Patch Local Binary Patterns (TPLBP), Four-Patch Local Binary Patterns (FPLBP), Binarized Statistical Image Features (BSIF) and Local Phase Quantization (LPQ). Furthermore, experiments on the combinations of the four local descriptors at feature level using simply histograms concatenation are provided. The performance of the proposed approach is evaluated with different dimensionality reduction algorithms: Principal Component Analysis (PCA), Orthogonal Locality Preserving Projection (OLPP) and the combined PCA+EFM (Enhanced Fisher linear discriminate Model). Finally, multi-class Support Vector Machine (SVM) is used as a classifier to carry out the verification between imposters and customers. The proposed method has been tested on CASIA-3D face database and the experimental results show that our method achieves a high verification performance.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.52 no.1
• /
• pp.45-50
• /
• 2003

Texture analysis has been efficiently utilized in the area of terrain classification. In this application features have been obtained in the 2D image domain. This paper suggests 3D co-occurrence texture features by extending the concept of co-occurrence to 3D world. The suggested 3D features are described using co-occurrence histogram of digital elevations at two contiguous position as co-occurrence matrix. The practical construction of co-occurrence matrix limits the number of levels of digital elevation. If the digital elevation is quantized into the number of levels over the whole DEM(Digital Elevation Map), the distinctive features can not be obtained. To resolve the quantization problem, we employ local quantization technique which preserves the variation of elevations. Experiments has been carried out to verify the proposed 3D co-occurrence features, and the addition of the suggested features significantly improves the classification accuracy.

• Journal of Korea Multimedia Society
• /
• v.23 no.12
• /
• pp.1540-1551
• /
• 2020

Aiming at the problem that the existing human behavior recognition algorithm cannot fully utilize the multi-level spatio-temporal information of the network, a human behavior recognition algorithm based on a dense three-dimensional residual network is proposed. First, the proposed algorithm uses a dense block of three-dimensional residuals as the basic module of the network. The module extracts the hierarchical features of human behavior through densely connected convolutional layers; Secondly, the local feature aggregation adaptive method is used to learn the local dense features of human behavior; Then, the residual connection module is applied to promote the flow of feature information and reduced the difficulty of training; Finally, the multi-layer local feature extraction of the network is realized by cascading multiple three-dimensional residual dense blocks, and use the global feature aggregation adaptive method to learn the features of all network layers to realize human behavior recognition. A large number of experimental results on benchmark datasets KTH show that the recognition rate (top-l accuracy) of the proposed algorithm reaches 93.52%. Compared with the three-dimensional convolutional neural network (C3D) algorithm, it has improved by 3.93 percentage points. The proposed algorithm framework has good robustness and transfer learning ability, and can effectively handle a variety of video behavior recognition tasks.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.3
• /
• pp.31-39
• /
• 2006

This paper presents a novel approach to real-time recognition of 3D environment and objects for various applications such as intelligent robots, intelligent vehicles, intelligent buildings,..etc. First, we establish the three fundamental principles that humans use for recognizing and interacting with the environment. These principles have led to the development of an integrated approach to real-time 3D recognition and modeling, as follows: 1) It starts with a rapid but approximate characterization of the geometric configuration of workspace by identifying global plane features. 2) It quickly recognizes known objects in environment and replaces them by their models in database based on 3D registration. 3) It models the geometric details the geometric details on the fly adaptively to the need of the given task based on a multi-resolution octree representation. SIFT features with their 3D position data, referred to here as stereo-sis SIFT, are used extensively, together with point clouds, for fast extraction of global plane features, for fast recognition of objects, for fast registration of scenes, as well as for overcoming incomplete and noisy nature of point clouds.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.54 no.3
• /
• pp.166-173
• /
• 2005

Preserving image features, edges and details in the process of impulsive noise filtering is an important problem. To avoid image blurring, only corrupted pixels must be filtered. In this paper, we propose an effective impulse noise detection method using Self-Organizing Neural Network(SONN) which applies median filter selectively for removing random-valued impulse noises while preserving image features, edges and details. Using a $3\times3$ window, we obtain useful local features with which impulse noise patterns are classified. SONN is trained with sample image patterns and each pixel pattern is classified by its local information in the image. The results of the experiments with various images which are the noise range of $5-15\%$ show that our method performs better than other methods which use multiple threshold values for impulse noise detection.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10b
• /
• pp.54-59
• /
• 1993

This paper presents a method of estimating 3-D surface geometrical features that are necessary for 3-D object recognition and image interpretation. The features, such as surface needle maps and curvatures, are computed from range or intensity images. In general, the range and intensity images are prone to noises, and hence the features computed by differentiation calculi on such a noisy image are hardly applicable to industrial recognition tasks. In our approach, we try to obtain a more accurate estimate of the features by using a least-squares minimization procedure subject to local curvature consistency constraints. The algorithm is robust with respect to noises and is completely independent of the viewpoint at which the image is taken. The performance of the ajgoritlim is evaluated using both synthetic data and real intensity images.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.301-303
• /
• 2003

The extraction of 3D road network from high-resolution aerial images is still one of the current challenges in digital photogrammetry and computer vision. For many years, there are many researcher groups working for this task, but unt il now, there are no papers for doing this with TLS (Three linear scanner), which has been developed for the past several years, and has very high-resolution (about 3 cm in ground resolution). In this paper, we present a methodology of road extraction from high-resolution digital imagery taken over urban areas using this modern photogrammetry’s scanner (TLS). The key features of the approach are: (1) Because of high resolution of TLS image, our extraction method is especially designed for constructing 3D road map for next -generation digital navigation map; (2) for extracting road, we use the global context of the intensity variations associated with different features of road (i.e. zebra line and center line), prior to any local edge. So extraction can become comparatively easy, because we can use different special edge detector according different features. The results achieved with our approach show that it is possible and economic to extract 3D road data from Three Linear Scanner to construct next -generation digital navigation road map.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.3
• /
• pp.332-338
• /
• 2003

Content-based image retrieval is the research of images from database, that are visually similar to given image examples. Gabor functions and Gabor filters are regarded as excellent methods for feature extraction and texture segmentation. However, they have a disadvantage not to perform well in case of a rotated image because of its direction-oriented filter. This paper proposes a method of extracting local texture features from blocks with central interest points detected in an image and a rotation invariant Gabor wavelet filter. We also propose a method of comparing pattern histograms of features classified by VQ (Vector Quantization) among images.

• Journal of information and communication convergence engineering
• /
• v.9 no.2
• /
• pp.207-211
• /
• 2011

This paper presents a method for facial feature representation and recognition from the Covariance Matrices of the Gabor-filtered images. Gabor filters are a very powerful tool for processing images that respond to different local orientations and wave numbers around points of interest, especially on the local features on the face. This is a very unique attribute needed to extract special features around the facial components like eyebrows, eyes, mouth and nose. The Covariance matrices computed on Gabor filtered faces are adopted as the feature representation for face recognition. Geodesic distance measure is used as a matching measure and is preferred for its global consistency over other methods. Geodesic measure takes into consideration the position of the data points in addition to the geometric structure of given face images. The proposed method is invariant and robust under rotation, pose, or boundary distortion. Tests run on random images and also on publicly available JAFFE and FRAV3D face recognition databases provide impressively high percentage of recognition.