• The Journal of Korea Robotics Society
• /
• v.19 no.1
• /
• pp.98-105
• /
• 2024

This paper proposes a new approach with a distance-based regularization to the entropy applied to the NBV (Next-Best-View) selection with NeRF (Neural Radiance Fields). 3D reconstruction requires images from various viewpoints, and selecting where to capture these images is a highly complex problem. In a recent work, image acquisition was derived using NeRF's ray-based uncertainty. While this work was effective for evaluating candidate viewpoints at fixed distances from a camera to an object, it is limited when dealing with a range of candidate viewpoints at various distances, because it tends to favor selecting viewpoints at closer distances. Acquiring images from nearby viewpoints is beneficial for capturing surface details. However, with the limited number of images, its image selection is less overlapped and less frequently observed, so its reconstructed result is sensitive to noise and contains undesired artifacts. We propose a method that incorporates distance-based regularization into entropy, allowing us to acquire images at distances conducive to capturing both surface details without undesired noise and artifacts. Our experiments with synthetic images demonstrated that NeRF models with the proposed distance and entropy-based criteria achieved around 50 percent fewer reconstruction errors than the recent work.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.4
• /
• pp.9-15
• /
• 2016

Symmetry is everywhere in the world around us from galaxy to microbes. From ancient times symmetry is considered to be a reflection of the harmony of universe. Symmetry is not only a significant clue for human cognitive process, but also useful information for computer vision such as image understanding system. Application areas include face detection and recognition, indexing of image database, image segmentation and detection, analysis of medical images, and so on. The technique used in this paper extracts edges, and the perpendicular bisector of any two edge points is considered to be a candidate axis of symmetry. The coefficients of candidate axis are accumulated in the coefficient space. Then the axis of symmetry is determined to be the line for which the coefficient histogram has maximum value. In this paper, an improved method is proposed that utilizes the directional information of edges, which is a byproduct of the edge detection process. Experiment on 20 test images shows that the proposed method performs 22.7 times faster than the original method. In another test on 5 images with 4% salt-and-pepper noise, the proposed method detects the symmetry successfully, while the original method fails. This result reveals that the proposed method enhances the speed and accuracy of detection process at the same time.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.14 no.4
• /
• pp.332-339
• /
• 2014

Face classification has wide applications in security and surveillance. However, this technique presents various challenges caused by pose, illumination, and expression changes. Face recognition with long-distance images involves additional challenges, owing to focusing problems and motion blurring. Multiple frames under varying spatial or temporal settings can acquire additional information, which can be used to achieve improved classification performance. This study investigates the effectiveness of multi-frame decision-level fusion with photon-counting linear discriminant analysis. Multiple frames generate multiple scores for each class. The fusion process comprises three stages: score normalization, score validation, and score combination. Candidate scores are selected during the score validation process, after the scores are normalized. The score validation process removes bad scores that can degrade the final output. The selected candidate scores are combined using one of the following fusion rules: maximum, averaging, and majority voting. Degraded facial images are employed to demonstrate the robustness of multi-frame decision-level fusion in harsh environments. Out-of-focus and motion blurring point-spread functions are applied to the test images, to simulate long-distance acquisition. Experimental results with three facial data sets indicate the efficiency of the proposed decision-level fusion scheme.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.8
• /
• pp.137-147
• /
• 1994

In this paper,we propose a knowledge-based system for locating Destination Address Block(DAB) by analyzing the structure of Korean envelope images. In the proposed system the preprocessing steps such as adaptive binarization connected component extraction and deskewing are carried out first for the effective structure analysis of the envelope image. Then DAB containing address name and zipcode parts of the input envelope image is extracted by an iterative procedure based on the knowledge acquired from the statistical feature analysis of the various envelope images. Most of the system for slocating address blocks on envelopes have extracted DAB by segmenting an envelope image into several candidate blocks followed by selecting one among the candidate blocks. Because it is very difficult to segment a Korean envelope image into several blocks due to the specific writing habits that the addresses on the envelope are written in close proximity to each other the proposed iterative procedure determines DAB by splitting or merging the connected components and verifies the determined DAB without segmentation and selection. Experiments with a great number of the live envelopes provided from Seoul Mail Center in Koorea were carried out. The results reveal that the proposed system is very effective for address block location on Korean envelopes.

• 전자공학회논문지 IE
• /
• v.44 no.2
• /
• pp.32-40
• /
• 2007

In this paper, the crease extraction method is proposed to improve the accuracy of feature extraction within the fingerprint image. First of all, for each pixel in fingerprint image, it calculates the average grey level and variance to determine if the current pixel composes the crease, and estimates the direction of crease. Secondly, once the direction of every pixel in crease candidate area is estimated, it is decomposed into 8 different images, depending on their direction. The properties of crease consists of the length of the crease candidate area, the correspondence between the crease direction and the pixel distribution direction, the difference between the ridge direction and the pixel distribution direction, and finally the grey level of the candidate pixels. The proposed method finally extracts the crease from the crease clusters estimated from directional images. In conclusion, applying the proposed method improved the accuracy of overall feature extraction by 91.4% by accurately and precisely extracting the crease from fingerprint image.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.12
• /
• pp.3683-3693
• /
• 1999

Most recognition methods of car licence plate have difficulties concerning plate recognition rates and system stability in that restricted car images are used and good image capture environment is required. To overcome these difficulties, I proposed a new recognition method of car licence plates, in which both intensity variation and color information are used. For a captured car image, multiple candidate plate-bands are extracted based on the number of intensity variation. To have an equal performance on abnormally dark and bright Images. plate lightness is calculated and adjusted based on the brightness of plate background. Candidate plate regions are extracted using contour following on plate color pixels in oath plate band. A candidate region is decided as a real plate region after extracting character regions and then recognizing them. I recognize characters using template matching since total number of possible characters is small and they art machine printed. To show the efficiency of the proposed method, I tested it on 200 car images and found that the method shows good performance.

• Journal of radiological science and technology
• /
• v.41 no.3
• /
• pp.231-240
• /
• 2018

Automated breast ultrasound (ABUS) equipment is a new innovative technique for 3D automatic breast scanning, but limited for the examination in the concave axillary region. The purpose of this study was to determine feasible candidate materials for the ultrasonic wave propagation media in ABUS, enabling the evaluation of the axillary region. Ultrasonography was performed using an ABUS system ($Invenia^{TM}ABUS$, GE, USA) on the ultrasound-specific phantom (UC-551M-0.5, ATS Laboratories, USA) covered by different candidate materials. The validity of feasible candidate materials was evaluated by image quality. Three independent radiological technologists, with more than 10 years of experience, visually assessed on the images. The inter-observer agreements according to the candidate materials were tested using Cronbach's alpha. Unenveloped solidified carrageenan can be a feasible material for the use of ABUS with excellent test reliability. Therefore, the coverage of the axillary region with carrageenan may be effective for ABUS which was originally developed for the convex anatomic structure as female breast.

• Journal of Broadcast Engineering
• /
• v.22 no.5
• /
• pp.643-651
• /
• 2017

Steganography is one of information hiding technologies and discriminated from cryptography in that it focuses on avoiding the existence the hidden information from being detected by third parties, rather than protecting it from being decoded. In this paper, as an image steganography method which uses images as media, we propose a new block matching method that embeds information into the discrete wavelet transform (DWT) domain. The proposed method, based on a statistical analysis, reduces loss of embedding capacity due to inequable use of candidate blocks. It works in such a way that computes the variance of each candidate block, preserves candidate blocks with high frequency components while reducing candidate blocks with low frequency components by compressing them exploiting the k-means clustering algorithm. Compared with the previous block matching method, the proposed method can reconstruct secret images with similar PSNRs while embedding higher-capacity information.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.2
• /
• pp.885-892
• /
• 2013

In this paper, we propose a method of effectively segmenting character regions by using texture and depth features in 3D stereoscopic images. The suggested method is mainly composed of four steps. The candidate character region extraction step extracts candidate character regions by using texture features. The character region localization step obtains only the string regions in the candidate character regions. The character/background separation step separates characters from background in the localized character areas. The verification step verifies if the candidate regions are real characters or not. In experimental results, we show that the proposed method can extract character regions from input images more accurately compared to other existing methods.

• Journal of KIISE:Software and Applications
• /
• v.37 no.4
• /
• pp.264-272
• /
• 2010

Example-based super resolution is a method to restore a high resolution image from low resolution images through training and retrieval of image patches. It is not only good in its performance but also available for a single frame low-resolution image. However, its time complexity is very high because it requires lots of comparisons to retrieve image patches in restoration process. In order to improve the restoration speed, an efficient patch retrieval algorithm is essential. In this paper, we applied various high-dimensional feature retrieval methods, available for the patch retrieval, to a practical example-based super resolution system and compared their speed. As well, we propose to apply the multi-phase candidate reduction approach to the patch retrieval process, which was successfully applied in character recognition fields but not used for the super resolution. In the experiments, LSH was the fastest among conventional methods. The multi-phase candidate reduction method, proposed in this paper, was even faster than LSH: For $1024{\times}1024$ images, it was 3.12 times faster than LSH.