• International journal of advanced smart convergence
• /
• v.5 no.3
• /
• pp.1-7
• /
• 2016

In existing Convolutional Neural Network (CNNs) for object recognition task, there are only few efforts known to reduce the noises from the images. Both convolution and pooling layers perform the features extraction without considering the noises of the input image, treating all pixels equally important. In computer vision field, there has been a study to weight a pixel importance. Seam carving resizes an image by sacrificing the least important pixels, leaving only the most important ones. We propose a new way to combine seam carving approach with current existing CNN model for object recognition task. We attempt to remove the noises or the "unimportant" pixels in the image before doing convolution and pooling, in order to get better feature representatives. Our model shows promising result with CIFAR-10 dataset.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.11 no.2
• /
• pp.175-180
• /
• 2018

In this paper, a feature extraction and object recognition algorithm using only morphological operations is proposed. The morphological operations used in feature extraction are erosion and dilation, opening and closing combining erosion and dilation, and morphological edge and skeleton detection operation. In the process of recognizing an object based on features, a pooling operation is applied to reduce the dimension. Among various structuring elements, $3{\times}3$ rhombus, $3{\times}3$ square, and $5{\times}5$ circle are arbitrarily selected in morphological operation process. It has confirmed that the proposed algorithm can be applied in object recognition fields through experiments using Internet images.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.1
• /
• pp.41-49
• /
• 2023

A convolutional neural network (CNN) is a representative algorithm for implementing artificial neural networks. CNNs have improved on the issues of rapid increase in calculation amount and low object classification rates, which are associated with a conventional multi-layered fully-connected neural network (FNN). However, because of the rapid development of IT devices, the maximum resolution of images captured by current smartphone and tablet cameras has reached 108 million pixels (MP). Specifically, a traditional CNN algorithm requires a significant cost and time to learn and process simple, high-resolution images. Therefore, this study proposes an improved CNN algorithm for implementing an object classification learning model for simple, high-resolution images. The proposed method alters the adjacency matrix value of the pooling layer's max pooling operation for the CNN algorithm to reduce the high-resolution image learning model's creation time. This study implemented a learning model capable of processing 4, 8, and 12 MP high-resolution images for each altered matrix value. The performance evaluation result showed that the creation time of the learning model implemented with the proposed algorithm decreased by 36.26% for 12 MP images. Compared to the conventional model, the proposed learning model's object recognition accuracy and loss rate were less than 1%, which is within the acceptable error range. Practical verification is necessary through future studies by implementing a learning model with more varied image types and a larger amount of image data than those used in this study.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.01a
• /
• pp.77-78
• /
• 2019

게임엔진을 이용한 FPS 디펜스 게임은 유니티3D 엔진을 사용하여 개발 하였으며 1인칭 시점으로 제한시간동안 몰려오는 적군을 막아내며 목표물을 지키는 게임이다. 많은 오브젝트를 관리하기 위해서 오브젝트 풀링을 사용하여 오브젝트가 생성-제거의 반복시 메모리에 부담을 주게되는 것을 씬 시작시 가용할 오브젝트를 불러온 뒤에 필요시에만 사용 하는 방법으로 메모리의 부담을 적게 하였고 플레이 기록을 랭킹으로 하여 사용자 간에 경쟁심을 유발 할 수 있도록 하였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.113-116
• /
• 2021

일반적으로 딥러닝 기반의 객체 검출(Object Detection)기법은 합성곱 신경망(Convolutional Neural Network, CNN)을 통해 입력된 영상의 특징(Feature)을 추출하여 이를 통해 객체 검출을 수행한다. 최근 자연어 처리 분야에서 획기적인 성능을 보인 트랜스포머(Transformer)가 영상 분류, 객체 검출과 같은 컴퓨터 비전 작업을 수행하는데 있어 경쟁력이 있음이 드러나고 있다. 본 논문에서는 YOLOv4-CSP의 CSP 블록을 개선한 one-stage 방식의 객체 검출 네트워크를 제안한다. 개선된 CSP 블록은 트랜스포머(Transformer)의 멀티 헤드 어텐션(Multi-Head Attention)과 CSP 형태의 공간 피라미드 풀링(Spatial Pyramid Pooling, SPP) 연산을 기반으로 네트워크의 Backbone과 Neck에서의 feature 학습을 돕는다. 본 실험은 MSCOCO test-dev2017 데이터 셋으로 평가하였으며 제안하는 네트워크는 YOLOv4-CSP의 경량화 모델인 YOLOv4s-mish에 대하여 평균 정밀도(Average Precision, AP)기준 2.7% 향상된 검출 정확도를 보인다.

• Journal of Korea Game Society
• /
• v.9 no.6
• /
• pp.163-170
• /
• 2009

There is a request from the client, we almost apply dynamic memory allocating method using Accept() of looping method; thus, there could be process of connecting synchronously lots of client in most of On-line gaming server engine. However, this kind of method causes on-line gaming server which need to support and process the clients, longer loading and bottle necking. Therefore we propose the object pooling scheme to minimize the memory fragmentation and the load of the initialization to the client using an AcceptEx() and static allocating method for an efficient gaming server of the On-line in this paper. We design and implement the gaming server applying to our proposed scheme. Also, we show efficiency of our proposed scheme by performance analysis in this paper.

• The Journal of the Acoustical Society of Korea
• /
• v.32 no.5
• /
• pp.438-445
• /
• 2013

In this paper, we propose an enhancement of Japanese acoustic model which is trained with Korean speech database by using several combination strategies. We describe the strategies for training more than two language combination, which are Cross-Language Transfer, Cross-Language Adaptation, and Data Pooling Approach. We simulated those strategies and found a proper method for our current Japanese database. Existing combination strategies are generally verified for under-resourced Language environments, but when the speech database is not fully under-resourced, those strategies have been confirmed inappropriate. We made tyied-list with only object-language on Data Pooling Approach training process. As the result, we found the ERR of the acoustic model to be 12.8 %.

• Smart Media Journal
• /
• v.10 no.4
• /
• pp.80-90
• /
• 2021

Using optical music recognition technology based on deep learning, we propose to apply the results derived to VR games. To detect the music objects in the music sheet, the deep learning model used YOLO v5, and Hough transform was employed to detect undetected objects, modifying the size of the staff. It analyzes and uses BPM, maximum number of combos, and musical notes in VR games using output result files, and prevents the backlog of notes through Object Pooling technology for resource management. In this paper, VR games can be produced with music elements derived from optical music recognition technology to expand the utilization of optical music recognition along with providing VR contents.

• Journal of Broadcast Engineering
• /
• v.24 no.5
• /
• pp.827-835
• /
• 2019

In the pooling stage of a full reference image quality assessment (FR-IQA) technique, the global perceived quality for any distorted image is usually measured from the quality of its local image patches. But all the image patches do not have equal contribution when estimating the overall visual quality since the degree of degradation on those patches depends on various considerations i.e., types of the patches, types of the distortions, distortion sensitivities of the patches, saliency score of the patches, etc. As a result, weighted pooling strategy comes into account and different weighting mechanisms are used by the existing FR-IQA methods. This paper performs a thorough analysis and proposes a novel weighting function by considering the luminance adaptation as well as the visual saliency effect to offer more appropriate local weights, which can be adopted in the existing FR-IQA frameworks to improve their prediction accuracy. The extended experimental results show the effectiveness of the proposed weighting function.

• Journal of Korea Multimedia Society
• /
• v.21 no.12
• /
• pp.1425-1430
• /
• 2018

In this paper, we propose a CNN based deep learning algorithm for semantic segmentation of images. In order to improve the accuracy of semantic segmentation, we combined pixel level object classification and image level object classification. The image level object classification is used to accurately detect the characteristics of an image, and the pixel level object classification is used to indicate which object area is included in each pixel. The proposed network structure consists of three parts in total. A part for extracting the features of the image, a part for outputting the final result in the resolution size of the original image, and a part for performing the image level object classification. Loss functions exist for image level and pixel level classification, respectively. Image-level object classification uses KL-Divergence and pixel level object classification uses cross-entropy. In addition, it combines the layer of the resolution of the network extracting the features and the network of the resolution to secure the position information of the lost feature and the information of the boundary of the object due to the pooling operation.