• Journal of Convergence for Information Technology
• /
• v.7 no.2
• /
• pp.31-36
• /
• 2017

In this paper, we try to search and reproduce the image part of a specific cast from a large number of images. The conventional method must manually set the offset value when searching for a scene or viewing a corner. However, in this paper, the proposed method learns the main character 's face, then finds the main character in the image recognition and moves to the scene where the main character appears to reproduce the image. Data for specific performers is extracted and collected using crawl techniques. Based on the collected data, we learn using convolutional neural network algorithm and perform performance evaluation using it. The performance evaluation measures the accuracy by extracting and judging a specific performer learned in the extracted key frame while playing the drama. The performance confirmation of how quickly and accurately the learned scene is searched has obtained about 93% accuracy. Based on the derived performance, it is applied to the image service such as viewing, searching for person and detailed information retrieval per corner

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.06a
• /
• pp.102-104
• /
• 2017

최근 Convolutional neural network (CNN)을 도입하여, SAR 영상의 목표물 인식 알고리즘이 높은 성능을 보여주었다. SAR 영상은 4 종류의 polarization 정보로 구성되어있다. 기계와 신호처리의 비용으로 인하여 일부 데이터는 적은 수의 polarization 정보를 가지고 있다. 따라서 우리는 SAR 영상 data 를 멀티모달 데이터로 해석하였다. 그리고 우리는 이러한 멀티모달 데이터에 잘 작동할 수 있는 콘볼루션 신경망을 제안하였다. 우리는 데이터가 포함하는 모달의 수에 반비례 하도록 scale factor 구성하고 이를 입력 크기조절에 사용하였다. 입력의 크기를 조절하여, 네트워크는 특징맵의 크기를 모달의 수와 상관없이 일정하게 유지할 수 있었다. 또한 제안하는 입력 크기조절 방법은 네트워크의 dead filter 의 수를 감소 시켰고, 이는 네트워크가 자신의 capacity 를 잘 활용한다는 것을 의미한다. 또 제안된 네트워크는 특징맵을 구성할 때 다양한 모달을 활용하였고, 이는 네트워크가 모달간의 상관관계를 학습했다는 것을 의미한다. 그 결과, 제안된 네트워크의 성능은 입력 크기조절이 없는 일반적인 네트워크보다 높은 성능을 보여주었다. 또한 우리는 전이학습의 개념을 이용하여 네트워크를 모달의 수가 많은 데이터부터 차례대로 학습시켰다. 전이학습을 통하여 네트워크가 학습되었을 때, 제안된 네트워크는 특정 모달의 조합 경우만을 위해 학습된 네트워크보다 높은 성능을 보여준다.

• 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.

• Journal of Sensor Science and Technology
• /
• v.33 no.2
• /
• pp.98-104
• /
• 2024

This study proposes a method to overcome the limited detection range of short-baseline stereo cameras (SBSCs). The proposed method includes two steps: (1) predicting an unscaled initial depth using monocular depth estimation (MDE) and (2) adjusting the unscaled initial depth by a scale factor. The scale factor is computed by triangulating the sparse visual keypoints extracted from the left and right images of the SBSC. The proposed method allows the use of any pre-trained MDE model without the need for additional training or data collection, making it efficient even when considering the computational constraints of small platforms. Using an open dataset, the performance of the proposed method was demonstrated by comparing it with other conventional stereo-based depth estimation methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.3
• /
• pp.389-395
• /
• 2021

In this paper, we propose a target image exchange model to improve performance of the object tracking algorithm based on a Siamese network. The object tracking algorithm based on the Siamese network tracks the object by finding the most similar part in the search image using only the target image specified in the first frame of the sequence. Since only the object of the first frame and the search image compare similarity, if tracking fails once, errors accumulate and drift in a part other than the tracked object occurs. Therefore, by designing a CNN(Convolutional Neural Network) based model, we check whether the tracking is progressing well, and the target image exchange timing is defined by using the score output from the Siamese network-based object tracking algorithm. The proposed model is evaluated the performance using the VOT-2018 dataset, and finally achieved an accuracy of 0.611 and a robustness of 22.816.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.17 no.4
• /
• pp.138-149
• /
• 2018

In this paper, we propose a system model which effectively mitigates impulsive noise that degrades the performance of power line communication. Recently, deep learning have shown effective performance improvement in various fields. In order to mitigate effective impulsive noise, we applied a convolution neural network which is one of deep learning algorithm to conventional system. Also, we used a successive interference cancellation scheme to mitigate impulsive noise generated from multi-users. We simulate the proposed model which can be applied to the power line communication in the Section V. The performance of the proposed system model is verified through bit error probability versus SNR graph. In addition, we compare ZF and MMSE successive interference cancellation scheme, successive interference cancellation with optimal ordering, and successive interference cancellation without optimal ordering. Then we confirm which schemes have better performance.

• Journal of Broadcast Engineering
• /
• v.24 no.2
• /
• pp.234-242
• /
• 2019

In this paper, we propose a deep neural network which extracts SIFT feature points by determining whether the center pixel of a cropped image is a SIFT feature point. The data set of this network consists of a DIV2K dataset cut into $33{\times}33$ size and uses RGB image unlike SIFT which uses black and white image. The ground truth consists of the RobHess SIFT features extracted by setting the octave (scale) to 0, the sigma to 1.6, and the intervals to 3. Based on the VGG-16, we construct an increasingly deep network of 13 to 23 and 33 convolution layers, and experiment with changing the method of increasing the image scale. The result of using the sigmoid function as the activation function of the output layer is compared with the result using the softmax function. Experimental results show that the proposed network not only has more than 99% extraction accuracy but also has high extraction repeatability for distorted images.

• Journal of Korea Multimedia Society
• /
• v.20 no.11
• /
• pp.1750-1758
• /
• 2017

In super-resolution, various methods with Convolutional Neural Network(CNN) have recently been proposed. CNN based methods provide much higher image quality than conventional methods. Especially, VDSR outperforms other CNN based methods in terms of image quality. However, it requires a high computational complexity which prevents real-time processing. In this paper, the method to apply a deconvolution layer to VDSR is proposed to reduce computational complexity. Compared to original VDSR, the proposed method achieves the 4.46 times speed-up and its degradation in image quality is less than -0.1 dB which is negligible.

• Journal of IKEEE
• /
• v.21 no.3
• /
• pp.244-247
• /
• 2017

In this paper, we implemented a convolution neural network using GP-GPU. After defining the structure, CNN performed inferencing using the GP-GPU with 256 threads, which was the previous study, using the weight obtained from the training. Training used Intel i7-4470 CPU and Matlab. Dataset used Daimler Pedestrian Dataset. The GP-GPU is controlled by the PC using PCIe and operates as an FPGA. We assigned a thread according to the depth and size of each layer. In the case of the pooling layer, we used over warpping pooling to perform additional operations on the horizontal and vertical regions. One inferencing takes about 12 ms.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1627-1634
• /
• 2021

This paper introduces a low-power compact ADC circuit for analog Convolutional filter for low-power neural network accelerator SOC. While convolutional neural network accelerators can speed up the learning and inference process, they have drawback of consuming excessive power and occupying large chip area due to large number of multiply-and-accumulate operators when implemented in complex digital circuits. To overcome these drawbacks, we implemented an analog convolutional filter that consists of an analog multiply-and-accumulate arithmetic circuit along with an ADC. This paper is focused on the design optimization of a low-power 8bit SAR ADC for the analog convolutional filter accelerator We demonstrate how to minimize the capacitor-array DAC, an important component of SAR ADC, which is three times smaller than the conventional circuit. The proposed ADC has been fabricated in CMOS 65nm process. It achieves an overall size of 1355.7㎛², power consumption of 2.6㎼ at a frequency of 100MHz, SNDR of 44.19 dB, and ENOB of 7.04bit.