• 한국멀티미디어학회논문지
• /
• 제23권2호
• /
• pp.135-145
• /
• 2020

Recently, deep-learning based methods for low-light image enhancement accomplish great success through supervised learning. However, they still suffer from the lack of sufficient training data due to difficulty of obtaining a large amount of low-/normal-light image pairs in real environments. In this paper, we propose an unsupervised learning approach for single low-light image enhancement using the bright channel prior (BCP), which gives the constraint that the brightest pixel in a small patch is likely to be close to 1. With this prior, pseudo ground-truth is first generated to establish an unsupervised loss function. The proposed enhancement network is then trained using the proposed unsupervised loss function. To the best of our knowledge, this is the first attempt that performs a low-light image enhancement through unsupervised learning. In addition, we introduce a self-attention map for preserving image details and naturalness in the enhanced result. We validate the proposed method on various public datasets, demonstrating that our method achieves competitive performance over state-of-the-arts.

• Journal of Information Processing Systems
• /
• 제10권3호
• /
• pp.443-458
• /
• 2014

An extreme learning machine (ELM) is a recently proposed learning algorithm for a single-layer feed forward neural network. In this paper we studied the ensemble of ELM by using a bagging algorithm for facial expression recognition (FER). Facial expression analysis is widely used in the behavior interpretation of emotions, for cognitive science, and social interactions. This paper presents a method for FER based on the histogram of orientation gradient (HOG) features using an ELM ensemble. First, the HOG features were extracted from the face image by dividing it into a number of small cells. A bagging algorithm was then used to construct many different bags of training data and each of them was trained by using separate ELMs. To recognize the expression of the input face image, HOG features were fed to each trained ELM and the results were combined by using a majority voting scheme. The ELM ensemble using bagging improves the generalized capability of the network significantly. The two available datasets (JAFFE and CK+) of facial expressions were used to evaluate the performance of the proposed classification system. Even the performance of individual ELM was smaller and the ELM ensemble using a bagging algorithm improved the recognition performance significantly.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권9호
• /
• pp.4476-4490
• /
• 2017

The aim of image annotation is to determine labels that can accurately describe the semantic information of images. Many approaches have been proposed to automate the image annotation task while achieving good performance. However, in most cases, the semantic similarities of images are ignored. Towards this end, we propose a novel Visual-Semantic Nearest Neighbor (VS-KNN) method by collectively exploring visual and semantic similarities for image annotation. First, for each label, visual nearest neighbors of a given test image are constructed from training images associated with this label. Second, each neighboring subset is determined by mining the semantic similarity and the visual similarity. Finally, the relevance between the images and labels is determined based on maximum a posteriori estimation. Extensive experiments were conducted using three widely used image datasets. The experimental results show the effectiveness of the proposed method in comparison with state-of-the-arts methods.

• 인터넷정보학회논문지
• /
• 제20권6호
• /
• pp.29-36
• /
• 2019

When dealing with outdoor images in a variety of computer vision applications, the presence of shadow degrades performance. In order to understand the information occluded by shadow, it is essential to remove the shadow. To solve this problem, in many studies, involves a two-step process of shadow detection and removal. However, the field of shadow detection based on CNN has greatly improved, but the field of shadow removal has been difficult because it needs to be restored after removing the shadow. In this paper, it is assumed that shadow is detected, and shadow-less image is generated by using original image and shadow mask. In previous methods, based on CGAN, the image created by the generator was learned from only the aspect of the image patch in the adversarial learning through the discriminator. In the contrast, we propose a novel method using a discriminator that judges both the whole image and the local patch at the same time. We not only use the residual generator to produce high quality images, but we also use joint loss, which combines reconstruction loss and GAN loss for training stability. To evaluate our approach, we used an ISTD datasets consisting of a single image. The images generated by our approach show sharp and restored detailed information compared to previous methods.

• Nuclear Engineering and Technology
• /
• 제52권11호
• /
• pp.2678-2685
• /
• 2020

The pipe bends and elbows in nuclear power plants (NPPs) are vulnerable to degradation mechanisms and can cause wall-thinning defects. As it is difficult to detect both the defects generated inside the wall-thinned pipes and the preliminary signs, the wall-thinning defects should be accurately estimated to maintain the integrity of NPPs. This paper proposes a deep fuzzy neural network (DFNN) method and estimates the collapse moment of wall-thinned pipe bends and elbows. The proposed model has a simplified structure in which the fuzzy neural network module is repeatedly connected, and it is optimized using the least squares method and genetic algorithm. Numerical data obtained through simulations on the pipe bends and elbows with extrados, intrados, and crown defects were applied to the DFNN model to estimate the collapse moment. The acquired databases were divided into training, optimization, and test datasets and used to train and verify the estimation model. Consequently, the relative root mean square (RMS) errors of the estimated collapse moment at all the defect locations were within 0.25% for the test data. Such a low RMS error indicates that the DFNN model is accurate in estimating the collapse moment for wall-thinned pipe bends and elbows.

• 지능정보연구
• /
• 제18권2호
• /
• pp.131-141
• /
• 2012

논문에서는 협업필터링(collaborative filtering : CF) 기반한 추천시스템의 정확도를 높일 수 있는 방법을 제안하고 그 효과를 분석한다. 일반적인 CF기반 추천시스템에서는 시스템 세팅(참조집단 크기, 유의도 수준 등)을 한 가지 정해서 모든 경우에 대해서 동일하게 적용한다. 본 논문에서는 개별 사용자의 특성에 따라 이러한 세팅을 최적화 해서 개별적으로 적용하는 방법을 개발하였다. 이런 개인화된 세팅의 효과를 측정하기 위해서 Netflix의 자료를 사용해서 일반적인 추천시스템과 추천 정확도를 비교하였다. 분석 결과, 동일한 세팅을 적용하는 일반적인 추천시스템에 비해서 개인화된 세팅을 적용한 경우 정확도가 월등히 향상됨을 확인하였다. 이 결과의 시사점과 함께 미래 연구의 방향에 대해서도 논의한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권1호
• /
• pp.249-268
• /
• 2014

Relevance feedback is an effective tool to bridge the gap between superficial image contents and medically-relevant sense in content-based medical image retrieval. In this paper, we propose an interactive medical image search framework based on pairwise constraint propagation. The basic idea is to obtain pairwise constraints from user feedback and propagate them to the entire image set to reconstruct the similarity matrix, and then rank medical images on this new manifold. In contrast to most of the algorithms that only concern manifold structure, the proposed method integrates pairwise constraint information in a feedback procedure and resolves the small sample size and the asymmetrical training typically in relevance feedback. We also introduce a long-term feedback strategy for our retrieval tasks. Experiments on two medical image datasets indicate the proposed approach can significantly improve the performance of medical image retrieval. The experiments also indicate that the proposed approach outperforms previous relevance feedback models.

• Nuclear Engineering and Technology
• /
• 제53권10호
• /
• pp.3431-3437
• /
• 2021

In this study, one-dimensional gamma ray source positions are estimated using a plastic scintillating optical fiber, two photon counters and via data processing with a machine learning algorithm. A nonlinear regression algorithm is used to construct a machine learning model for the position estimation of radioactive sources. The position estimation results of radioactive sources using machine learning are compared with the theoretical position estimation results based on the same measured data. Various tests at the source positions are conducted to determine the improvement in the accuracy of source position estimation. In addition, an evaluation is performed to compare the change in accuracy when varying the number of training datasets. The proposed one-dimensional gamma ray source position estimation system with plastic scintillating fiber using machine learning algorithm can be used as radioactive leakage scanners at disposal sites.

• 로봇학회논문지
• /
• 제14권1호
• /
• pp.31-39
• /
• 2019

A robot usually adopts ANN (artificial neural network)-based object detection and instance segmentation algorithms to recognize objects but creating datasets for these algorithms requires high labeling costs because the dataset should be manually labeled. In order to lower the labeling cost, a new scheme is proposed that can automatically generate a training images and label them for specific objects. This scheme uses an instance segmentation algorithm trained to give the masks of unknown objects, so that they can be obtained in a simple environment. The RGB images of objects can be obtained by using these masks, and it is necessary to label the classes of objects through a human supervision. After obtaining object images, they are synthesized with various background images to create new images. Labeling the synthesized images is performed automatically using the masks and previously input object classes. In addition, human intervention is further reduced by using the robot arm to collect object images. The experiments show that the performance of instance segmentation trained through the proposed method is equivalent to that of the real dataset and that the time required to generate the dataset can be significantly reduced.

• 한국멀티미디어학회논문지
• /
• 제22권1호
• /
• pp.1-9
• /
• 2019

Single image rain removal is a typical inverse problem which decomposes the image into a background scene and a rain streak. Recent works have witnessed a substantial progress on the task due to the development of convolutional neural network (CNN). However, existing CNN-based approaches train the network with synthetically generated training examples. These data tend to make the network bias to the synthetic scenes. In this paper, we present an unsupervised framework for removing rain streaks from real-world rainy images. We focus on the natural phenomena that static rainy scenes capture a common background but different rain streak. From this observation, we train siamese network with the real rain image pairs, which outputs identical backgrounds from the pairs. To train our network, a real rainy dataset is constructed via web-crawling. We show that our unsupervised framework outperforms the recent CNN-based approaches, which are trained by supervised manner. Experimental results demonstrate that the effectiveness of our framework on both synthetic and real-world datasets, showing improved performance over previous approaches.