• Journal of IKEEE
• /
• v.24 no.2
• /
• pp.507-513
• /
• 2020

The median filter that is used in various fields requiring image processing converts to a median value of pixels belonging to a radius r for all pixels in the image of n×m size. For 8-bit depth images, an O(nm) time algorithm exists but for over 16-bit depth images, there is an O(nmlog²r) time algorithm of Gil and Werman. In this paper, we propose an efficient median filter algorithm that works for more than 16-bit depth images. The time complexity of our algorithm is the same as that of Gil and Werman, but theoretical analysis and experimental results show that ours is efficient than above two times.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.15 no.4
• /
• pp.1-14
• /
• 2012

A river is defined as the watercourse flowing through its channel, and the mapping tasks of a river plays an important role for the research on the topographic changes in the riparian zones and the research on the monitoring of flooding in its floodplain. However, the utilization of the ground surveying technologies is not efficient for the mapping tasks of a river due to the irregular surfaces of the riparian zones and the dynamic changes of water level of a river. Recently, the spatial information data sets are widely used for the coastal mapping tasks due to the acquisition of the topographic information without human accessibility. In this research, we tried to extract a river from the RapidEye imagery by using the ISODATA(Iterative Self_Organizing Data Analysis) classification algorithm with the two different parameters(NIR (Near Infra-Red) band and NDVI(Normalized Difference Vegetation Index)). First, the two different images(the NIR band image and the NDVI image) were generated from the RapidEye imagery. Second, the ISODATA algorithm were applied to each image and each river was generated in each image through the post-processing steps. River boundaries were also extracted from each classified image using the Sobel edge detection algorithm. Ground truths determined by the experienced expert are used for the assessment of the accuracy of an each generated river. Statistical results show that the extracted river using the NIR band has higher accuracies than the extracted river using the NDVI.

• Journal of Biosystems Engineering
• /
• v.36 no.1
• /
• pp.40-47
• /
• 2011

Image analysis algorithm for the quality evaluation of ginseng seedling was investigated. The images of ginseng seedling were acquired with a color CCD camera and processed with the image analysis methods, such as binary conversion, labeling, and thinning. The processed images were used to calculate the length and weight of ginseng seedlings. The length and weight of the samples could be predicted with standard errors of 0.343 mm, and 0.0214 g respectively, $R^2$ values of 0.8738 and 0.9835 respectively. For the evaluation of the three quality grades of Gab, Eul, and abnormal ginseng seedlings, features from the processed images were extracted. The features combined with the ratio of the lengths and areas of the ginseng seedlings efficiently differentiate the abnormal shapes from the normal ones of the samples. The grade levels were evaluated with an efficient pattern recognition method of support vector machine analysis. The quality grade of ginseng seedling could be evaluated with an accuracy of 95% and 97% for training and validation, respectively. The result indicates that color image analysis with support vector machine algorithm has good potential to be used for the development of an automatic sorting system for ginseng seedling.

• Journal of Korea Multimedia Society
• /
• v.6 no.2
• /
• pp.247-258
• /
• 2003

Principal component analysis (PCA) based algorithms form the basis of numerous algorithms and studies in the face recognition literature. PCA is a statistical technique and its incorporation into a face recognition system requires numerous design decisions. We explicitly take the design decisions by in-troducing a generic modular PCA-algorithm since some of these decision ate not documented in the literature We experiment with different implementations of each module, and evaluate the different im-plementations using the September 1996 FERET evaluation protocol (the do facto standard method for evaluating face recognition algorithms). We experiment with (1) changing the illumination normalization procedure; (2) studying effects on algorithm performance of compressing images using JPEG and wavelet compression algorithms; (3) varying the number of eigenvectors in the representation; and (4) changing the similarity measure in classification process. We perform two experiments. In the first experiment, we report performance results on the standard September 1996 FERET large gallery image sets. The result shows that empirical analysis of preprocessing, feature extraction, and matching performance is extremely important in order to produce optimized performance. In the second experiment, we examine variations in algorithm performance based on 100 randomly generated image sets (galleries) of the same size. The result shows that a reasonable threshold for measuring significant difference in performance for the classifiers is 0.10.

• Korean Journal of Remote Sensing
• /
• v.23 no.3
• /
• pp.211-219
• /
• 2007

Building extraction is one of the essential issues for 3D city modelling. In recent years, high-resolution satellite imagery has become widely available and it brings new methodology for urban mapping. In this paper, we have developed a semi-automatic algorithm to determine building heights from monoscopic high-resolution satellite data. The algorithm is based on the analysis of the projected shadow and actual shadow of a building. Once two roof comer points are measured manually, the algorithm detects (rectangular) roof boundary automatically. Then it estimates a building height automatically by projecting building shadow onto the image for a given building height, counting overlapping pixels between the projected shadow and actual shadow, and finding the height that maximizes the number of overlapping pixels. Once the height and roof boundary are available, the footprint and a 3D wireframe model of a building can be determined. The proposed algorithm is tested with IKONOS images over Deajeon city and the result is compared with the building height determined by stereo analysis. The accuracy of building height extraction is examined using standard error of estimate.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.1
• /
• pp.67-72
• /
• 2012

In this paper, we present an image enhancement method for atmospherically degraded images using atmospheric light and transmission based on color analysis. We first generate a normalized image using maximum value of each RGB color channel. Then, each atmospheric light is estimated from RGB color channel respectively by calculating reflectance of an image. We also, generate a transmission using gamma coefficients from the Y channel of the image. We can significantly enhance the visibility of an image by using the estimated atmospheric light and the transmission. The proposed algorithm can remove atmospheric degradation components better than existing techniques because the color prevents color distortion which is common problem of existing techniques. Experimental results demonstrate that the proposed algorithm can improve visibility be removing fog, smoke, and dust.

• Journal of the Korean earth science society
• /
• v.43 no.5
• /
• pp.579-590
• /
• 2022

In this study, we aimed to develop a Python-based outer-ring galaxy analysis algorithm according to the data science process. We assumed that the potential users are citizen scientists, including students and teachers. In the actual classification studies using real data of galaxies, a specialized software called IRAF is used, thereby limiting the general public's access to the software. Therefore, an image analysis algorithm was developed for the outer-ring galaxies as targets, which were compared with those of the previous research. The results of this study were compared with those of studies conducted using IRAF to verify the performance of the newly developed image analysis algorithm. Among the 69 outer-ring galaxies in the first test, 50 cases (72.5%) showed high agreement with the previous research. The remaining 19 cases (27.5%) showed differences that were caused by the presence of bright stars overlapped in the line of sight or weak brightness in the inner galaxy. To increase the usability of the finished product that has undergone a supplementary process, all used data, algorithms, Python code files, and user manuals were loaded in GitHub and made available as shared educational materials.

• Journal of Korea Multimedia Society
• /
• v.20 no.1
• /
• pp.43-50
• /
• 2017

According to living environment, the number of deaths due to thyroid diseases increased. In this paper, we proposed an algorithm for recognizing a thyroid detection using texture analysis based on shape, gray level co-occurrence matrix and gray level run length matrix. First of all, we segmented the region of interest (ROI) using active contour model algorithm. Then, we applied a total of 18 features (5 first order descriptors, 10 Gray level co-occurrence matrix features(GLCM), 2 Gray level run length matrix features and shape feature) to each thyroid region of interest. The extracted features are used as statistical analysis. Our results show that first order statistics (Skewness, Entropy, Energy, Smoothness), GLCM (Correlation, Contrast, Energy, Entropy, Difference variance, Difference Entropy, Homogeneity, Maximum Probability, Sum average, Sum entropy), GLRLM features and shape feature helped to distinguish thyroid benign and malignant. This algorithm will be helpful to diagnose of thyroid nodule on ultrasound images.

• Korean Journal of Applied Biomechanics
• /
• v.13 no.1
• /
• pp.1-12
• /
• 2003

The purpose of this study was to design the method of 3D space calibration to reduce RMSE by statistical analysis when using the DLT algorithm and control frame. Control frame for 3D space calibration was consist of $1{\times}3{\times}2m$ and 162 contort points adhere to it. For calculate of 3D coordination used two methods about 2D coordination on image frame, 2D coordinate on each image frame and mean coordination. The methods of statistical analysis used one-way ANOVA and T-test. Significant level was ${\alpha}=.05$. The compose of methods for reduce RMSE were as follow. 1. Use the control frame composed of 24-44 control points arranged equally. 2. When photographing, locate control frame to center of image plane(image frame) o. use the lens of a few distortion. 3. When calculate of 3D coordination, use mean of 2D coordinate obtainable from all image frames.

• Journal of Korea Multimedia Society
• /
• v.18 no.8
• /
• pp.901-908
• /
• 2015

Methods of detecting, describing, matching image features, like corners and blobs, have been actively studied as a fundamental step for image processing and computer vision applications. As one of feature description/matching methods, LLAH(Locally Likely Arrangement Hashing) describes image features based on the geometric relationship between their neighbors, and thus is suitable for scenes with poor texture. This paper presents a modified LLAH algorithm, which includes the image features themselves for robustly describing the geometric relationship unlike the original LLAH, and employes a voting-based feature matching scheme that makes feature description much simpler. Then, this paper quantitatively analyzes its performance with synthetic images in the presence of Gaussian noise.