• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.5A
• /
• pp.375-387
• /
• 2012

This paper proposes a new real-time object tracking algorithm using particle filters with color and texture features in moving CCD camera images. If the user selects an initial object, this region is declared as a target particle and an initial state is modeled. Then, N particles are generated based on random distribution and CS-LBP (Centre Symmetric Local Binary Patterns) for texture model and weighted color distribution is modeled from each particle. For observation likelihoods estimation, Bhattacharyya distance between particles and their feature models are calculated and this observation likelihoods are used for weights of individual particles. After weights estimation, a new particle which has the maximum weight is selected and new particles are re-sampled using the maximum particle. For performance comparison, we tested a few combinations of features and particle filters. The proposed algorithm showed best object tracking performance when we used color and texture model simultaneously for likelihood estimation.

• Korean Journal of Remote Sensing
• /
• v.35 no.6_2
• /
• pp.1037-1052
• /
• 2019

We had developed in preceding study a classification model for the Korean pine and Larch with an accuracy of 98 percent using Hyperion and Sentinel-2 satellite images, texture information, and geometric information as the first step for tree species mapping in the inaccessible North Korea. Considering a share of major tree species in North Korea, the classification model needs to be expanded as it has a large share of Oak(29.5%), Pine (12.7%), Fir (8.2%), and as well as Larch (17.5%) and Korean pine (5.8%). In order to classify 5 major tree species, national forest type map of South Korea was used to build 11,039 training and 2,330 validation data. Sentinel-2 data was used to derive spectral information, and PlanetScope data was used to generate texture information. Geometric information was built from SRTM DEM data. As a machine learning algorithm, Random forest was used. As a result, the overall accuracy of classification was 80% with 0.80 kappa statistics. Based on the training data and the classification model constructed through this study, we will extend the application to Mt. Baekdu and North and South Goseong areas to confirm the applicability of tree species classification on the Korean Peninsula.

• Korean Journal of Remote Sensing
• /
• v.37 no.2
• /
• pp.291-304
• /
• 2021

Updating a forest type map is essential for sustainable forest resource management and monitoring to cope with climate change and various environmental problems. According to the necessity of efficient and wide-area forestry remote sensing, CAS500-4 (Compact Advanced Satellite 500-4; The agriculture and forestry satellite) project has been confirmed and scheduled for launch in 2023. Before launching and utilizing CAS500-4, this study aimed to pre-evaluation the possibility of satellite-based tree species classification using RapidEye, which has similar specifications to the CAS500-4. In this study, the study area was the Chuncheon forest management complex, Gangwon-do. The spectral information was extracted from the growing season image. And the GLCM texture information was derived from the growing and non-growing seasons NIR bands. Both information were used to classification with random forest machine learning method. In this study, tree species were classified into nine classes to the coniferous tree (Korean red pine, Korean pine, Japanese larch), broad-leaved trees (Mongolian oak, Oriental cork oak, East Asian white birch, Korean Castanea, and other broad-leaved trees), and mixed forest. Finally, the classification accuracy was calculated by comparing the forest type map and classification results. As a result, the accuracy was 39.41% when only spectral information was used and 69.29% when both spectral information and texture information was used. For future study, the applicability of the CAS500-4 will be improved by substituting additional variables that more effectively reflect vegetation's ecological characteristics.

• Journal of information and communication convergence engineering
• /
• v.13 no.3
• /
• pp.205-214
• /
• 2015

In this paper, we propose a new approach to learning a discriminative model for figure/ground segmentation by incorporating the bag-of-features and conditional random field (CRF) techniques. We advocate the use of image patches instead of superpixels as the basic processing unit. The latter has a homogeneous appearance and adheres to object boundaries, while an image patch often contains more discriminative information (e.g., local image structure) to distinguish its categories. We use pixel-level sparse coding to represent an image patch. With the proposed feature representation, the unary classifier achieves a considerable binary segmentation performance. Further, we integrate unary and pairwise potentials into the CRF model to refine the segmentation results. The pairwise potentials include color and texture potentials with neighborhood interactions, and an edge potential. High segmentation accuracy is demonstrated on three benchmark datasets: the Weizmann horse dataset, the VOC2006 cow dataset, and the MSRC multiclass dataset. Extensive experiments show that the proposed approach performs favorably against the state-of-the-art approaches.

• Journal of the Korea Computer Graphics Society
• /
• v.6 no.3
• /
• pp.17-23
• /
• 2000

Recently, as it is growing concern about Non-Photo-Realistic Rendering, several researchers are studying to simulate conventional artistic expressions such as pen-and-ink, watercolor and impressionism techniques. In this paper, we describe a mosaic technique using color paper tore by hands. We will also try to represent real paper by using random midpoint displacement method (RMD) to express a shape of paper tore by hands without using cutters or scissors. The Perlin's noise function is used to express texture of paper.

• Journal of the Korean Society of Industry Convergence
• /
• v.23 no.3
• /
• pp.515-521
• /
• 2020

Deep learning methods have been effectively used to provide great improvement in various research fields such as machine learning, image processing and computer vision. One of the most frequently used deep learning methods in image processing is the convolutional neural networks. Compared to the traditional artificial neural networks, convolutional neural networks do not use the predefined kernels, but instead they learn data specific kernels. This property makes them to be used as feature extractors as well. In this study, we compared the quality of CNN features for traditional texture feature extraction methods. Experimental results demonstrate the superiority of the CNN features. Additionally, the recognition process and result of a pioneering CNN on MNIST database are presented.

• Macromolecular Research
• /
• v.9 no.2
• /
• pp.122-128
• /
• 2001

Monodisperse composite latex particles with size of ca. 300 nm, which consist ofn-butyl acrylate as a soft phase and methyl methacrylate as a hard phase with different morphology, were synthesized by seeded multi-stage emulsion polymerization. Three types of composite latex particles including random-, core/shell-, and gradient-type particles were obtained by using different monomer feeding methods during semi-batch emulsion polymerization. Effect of poly(butyl acrylate)-poly(methyl methacrylate) rubber particle morphology on the mechanical and rheological properties of rubber toughened poly(methyl methacrylate) was investigated. Among three different rubber particles, the gradient-type rubber particle showed better toughening effect than others. No significant variation of rheological property of poly(methyl methacrylate)/rubber blends was observed for the different rubber particle morphology.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.85-88
• /
• 2009

In this paper, we propose a DRF-based object detection method using the object adaptive patch in the satellite imagery. It is a Discriminative Random Fields (DRF) based work, so the detection is done by labeling to the possible patches in the image. For the feature information of each patch, we use the multi-scale and object adaptive patch and its texton histogram, instead of using the single scale and fixed grid patch. So, we can include contextual layout of texture information around the object. To make object adaptive patch, we use "superpixel lattice" scheme. As a result, each group of labeled patches represents the object or object's presence region. In the experiment, we compare the detection result with a fixed grid scheme and shows our result is more close to the object shape.

• Journal of the Korean institute of surface engineering
• /
• v.31 no.1
• /
• pp.24-33
• /
• 1998

Effects of pH and supporting salts on the characteristics of electrogalvanzied coating in sulfate bath are investigated. The fine grain size is obtained and the whiteness with the amount of supporting salts or pH increased at more than current density of 100A/$dm^2$<\TEX>, With supporting salts increased, the electro-conductivity of the bulk solution increases and the cell voltage decreases, while the width of the cathode burned edge gets wider because it seems that the increased overpotential the vicinity of cathode causes the decreases, of limiting current density. When the amount of supporting salts or pH of sulfate bath decreases, the zinc crystals have preferred orientation (001) planes. However when the amount of supporting salts or pH increase, the crystal texture has less (001) planes and gets to have random crystal planes.

• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.270-273
• /
• 2007

In this paper, a SAR-despeck1ing approach of adaptive iteration based a Bayesian model using the lognormal distribution for image intensity and a Gibbs random field (GRF) for image texture is proposed for noise removal of the images that are corrupted by multiplicative speckle noise. When the image intensity is logarithmically transformed, the speckle noise is approximately Gaussian additive noise, and it tends to a normal probability much faster than the intensity distribution. The MRF is incorporated into digital image analysis by viewing pixel types as states of molecules in a lattice-like physical system. The iterative approach based on MRF is very effective for the inner areas of regions in the observed scene, but may result in yielding false reconstruction around the boundaries due to using wrong information of adjacent regions with different characteristics. The proposed method suggests an adaptive approach using variable parameters depending on the location of reconstructed area, that is, how near to the boundary. The proximity of boundary is estimated by the statistics based on edge value, standard deviation, entropy, and the 4th moment of intensity distribution.