• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.2
• /
• pp.318-328
• /
• 2014

In this paper, as a way of performance improvement of business card recognition in the mobile environment, we suggested a hybrid OCR agent which combines data using a parallel processing sequence between various algorithms and different kinds of business card recognition engines which have learning data. We also suggested an Image Processing Method on mobile cameras which adapts to the changes of the lighting, exposing axis and the backgrounds of the cards which occur depending on the photographic conditions. In case a hybrid OCR agent is composed by the method suggested above, the average recognition rate of Korean business cards has improved from 90.69% to 95.5% compared to the cases where a single engine is used. By using the Image Processing Method, the image capacity has decreased to the average of 50%, and the recognition has improved from 83% to 92.48% showing 9.4% improvement.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.2
• /
• pp.187-192
• /
• 2021

This study proposes a system for take-off in a forest or similar complex environment using an object detector. In the simulator, a raspberry pi is mounted on a quadcopter with a length of 550mm between motors on a diagonal line, and the experiment is conducted based on edge computing. As for the images to be used for learning, about 150 images of 640⁎480 size were obtained by selecting three points inside Kunsan University, and then converting them to black and white, and pre-processing the binarization by placing a boundary value of 127. After that, we trained the SSD_Inception model. In the simulation, as a result of the experiment of taking off the drone through the model trained with the verification image as an input, a trajectory similar to the takeoff was drawn using the label.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.1
• /
• pp.67-75
• /
• 2024

Crosswalks play an important role for the safe movement of pedestrians in a complex urban environment. However, for the visually impaired, crosswalks can be a big risk factor. Although assistive tools such as braille blocks and acoustic traffic lights exist for safe walking, poor management can sometimes act as a hindrance to safety. This paper proposes a method to improve accuracy in a deep learning-based real-time crosswalk detection model that can be used in applications for pedestrian assistance for the disabled at the beginning. The image was binarized by utilizing the characteristic that the white line of the crosswalk image contrasts with the road surface, and through this, the crosswalk could be better recognized and the location of the crosswalk could be more accurately identified by using two models that learned the whole and the middle part of the crosswalk, respectively. In addition, it was intended to increase accuracy by creating a boundary box that recognizes crosswalks in two stages: whole and part. Through this method, additional frames that the detection model did not detect in RGB image learning from the crosswalk image could be detected.

• Investigative Magnetic Resonance Imaging
• /
• v.15 no.1
• /
• pp.32-40
• /
• 2011

Purpose : The purpose of this study was to investigate the microstructural changes according to aging on the thickness and signal intensity (SI) of the cortical gray matter (GM) and white matter (WM) on the T2-, fluid-attenuated inversion recovery (FLAIR) and T1-weighted MR images in normal subjects. Materials and Methods : The 10, 20, 30, 40, 50, 60, 70, 80 and 90 year age groups of men and women (each 10 individuals) who underwent routine brain MRI, including the T2-, FLAIR and T1-weighted images, were selected for this study. We measured the thickness and the SI of the cortical GM and WM at the postcentral gyrus, which has an even thickness at the level of centrum semiovale, on the axial scans and we calculated the mean values of the thickness ratio of the gray/white matter (TRGW) and the signal intensity ratio of the gray/white matter (SRGW), and we compared the ratios of each age group. Results : On the T2-weighted images, the TRGWs were 0.81 and 0.79 at the age of 10 and they were 0.73 and 0.71 at the age of 90 in the men and women, respectively. So, the GM thickness was decreased more than the WM thickness was with aging. On the FLAIR images, the TRGWs were 1.09 and 1.00 at the age of 10 and they were 1.11 and 0.95 at the age of 70 in the men and women, respectively. On the T1-weighted images, the TRGWs were 0.66 and 0.80 at the age of 10, and the ratio was changed to 0.90 and 0.78 at the age of 90 in the men and women, respectively. On the T2-weighted image, the SRGWs were 1.53 and 1.43 at the age of 10, and they were 1.23 and 1.27 at the age of 90 in the men and women, respectively. On the FLAIR images, the SRGWs were 1.23 and 1.25 at the age of 10 and they were 1.06 and 1.05 at the age of 90 in the men and women, respectively. On the T1-weighted images, the SRGWs were 0.86 and 0.85 at the age of 10, and they were 0.90 and 0.87 at the age of 90 in the men and women, respectively. Conclusion : We suggest that the age-related microstructural changes of the thickness and the SI of the cortical GM and WM on the T2-, FLAIR and T1-weighted images are unique, and so this knowledge will be helpful to differentiate neurodegenerative disease from normal aging of the brain.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.1
• /
• pp.43-50
• /
• 2016

This paper introduces an appearance-based facial expression recognition method using ASM landmarks which is used to acquire a detailed face region. In particular, EHMM-based algorithm and SVM classifier with histogram feature are employed to appearance-based facial expression recognition, and performance evaluation of proposed method was performed with CK and JAFFE facial expression database. In addition, performance comparison was achieved through comparison with distance-based face normalization method and a geometric feature-based facial expression approach which employed geometrical features of ASM landmarks and SVM algorithm. As a result, the proposed method using ASM-based face normalization showed performance improvements of 6.39% and 7.98% compared to previous distance-based face normalization method for CK database and JAFFE database, respectively. Also, the proposed method showed higher performance compared to geometric feature-based facial expression approach, and we confirmed an effectiveness of proposed method.

• Imaging Science in Dentistry
• /
• v.31 no.3
• /
• pp.181-184
• /
• 2001

Thirteen-year-old girl complaining of the swelling and pain on the left midface visited our dental hospital. On the radiographic examination, well-defined radiolucent lesion with hyperostotic border was found in the left maxilla accompanying with the external root resorption of the involved teeth and the displaced second molar. CT showed calcified bodies, thinning of hard palate, inferior orbital wall and lateral wall of nasal fossa, and thinning and perforation of the buccal plate of the maxilla. Enucleation and curettage of the lesion and nasoantrostomy was carried out and histopathologic examination mainly showed a solid tumor tissue composed of odontogenic epithelium and pulp tissues admixed with dentin and enamel formation. And some part of reduced follicular epithelium of tooth germ showed a change mimicking calcifying odontogenic cyst. Taken together, we concluded the lesion is an ameloblastic fibro-odontoma with a change of calcifying odontogenic cyst.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.8
• /
• pp.1433-1445
• /
• 1994

In this paper, an coherent optical correlator system based on the off-axis projection synthetic discriminant funtion (OA-pSDF) was analyzed and implemented optically. The filter was synthesized by combining conventional pSDF with single reference plane wave multiplexing. Synthesized pSDF were transformed to binary phase only filters (BPOFs) and fabricated as computer generated holograms(CGHs), which was used in the real time optical correlator system instead of using expensive spatial light modulators(SLMs). From the characteristic test, it was found that OA-pSDF showed distortion invariance and good performances in discrminating subset images. The proposed OA-pSDF BPOF could overcome the limitations of conventional BPOFs : that is distortion variance such as acale and rotation, especially out of plane variance.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.6
• /
• pp.545-555
• /
• 1991

In this paper, We discuss optical implementation of new optical adaptive patern recognition system based on single layer perception with learning capability and associative memory model having error corrective capability. The single layer perceptron is optically implemented by using 2 D LCTV spatial light modulators through the nonlinear quantization and polarization encoding methods, and 2 D hopfield associative memory is also implemented by using multifocus holographic lens. From some experimental results on classfication of Arabic numbers into even & edd numbers, it is shown that the proposed system can classify the patterns to the right classes correctly even for the partial and erronenous input patterns. Accordingly, the proposed optical adaptive pattern recognition system can be suggested for practical application in the fields of image processing and pattern recognition.

• Journal of Korea Entertainment Industry Association
• /
• v.14 no.6
• /
• pp.91-107
• /
• 2020

The purpose of this study is to explore quantitatively the success factors of YouTube's music cover channel. Along the process, constructing and testing the measurement for music cover channel's characteristics. The research constructed the measurement items based on results of qualitative studies. To perform the purposes the study suggest a research model which integrated the Post Acceptance Model with five music cover channel's characteristics based on the recent case studies. The five characteristics were contents, video, creator, interactivity, and publicity. The characteristics were integrated with Post Acceptance Model. After testing reliability and validity the research model was revised then, the research hypothesis was tested. The research results show: first, it was found that the content characteristics did not meet the expected value of users. Second, among the five factors of channel characteristics presented in creator's characteristics have the largest effect on user's value perception Third, in the factor analysis process, the content factor and the video were loaded on the same factor thus integrated as one factor. Fourth, satisfaction and intention for continual usage were also loaded on the same factor thus integrated as one factor. The research results suggest that music cover channel operators need to enhance the creator's attraction, communication with user, and publicity in various ways.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.13 no.6
• /
• pp.33-43
• /
• 2003

In this paper, we proposed a real-time watermarking algorithm to be combined and to work with a DWT(Discrete Wavelet Transform)-based image compressor. To reduce the amount of computation in selecting the watermarking positions, the proposed algorithm uses a pre-established look-up table for critical values, which was established statistically by computing the correlation according to the energy values of the corresponding wavelet coefficients. That is, watermark is embedded into the coefficients whose values are greater than the critical value in the look-up table which is searched on the basis of the energy values of the corresponding level-1 subband coefficients. Therefore, the proposed algorithm can operate in a real-time because the watermarking process operates in parallel with the compression procession without affecting the operation of the image compression. Also it improved the property of losing the watermark and the efficiency of image compression by watermark inserting, which results from the quantization and Huffman-Coding during the image compression. Visual recognizable patterns such as binary image were used as a watermark The experimental results showed that the proposed algorithm satisfied the properties of robustness and imperceptibility that are the major conditions of watermarking.