• Journal of information and communication convergence engineering
• /
• 제16권3호
• /
• pp.173-178
• /
• 2018

Accurate classification of cloud images is a challenging task. Almost all the existing methods rely on hand-crafted feature extraction. Their limitation is low discriminative power. In the recent years, deep learning with convolution neural networks (CNNs), which can auto extract features, has achieved promising results in many computer vision and image understanding fields. However, deep learning approaches usually need large datasets. This paper proposes a deep learning approach for classification of cloud image patches on small datasets. First, we design a suitable deep learning model for small datasets using a CNN, and then we apply data augmentation and dropout regularization techniques to increase the generalization of the model. The experiments for the proposed approach were performed on SWIMCAT small dataset with k-fold cross-validation. The experimental results demonstrated perfect classification accuracy for most classes on every fold, and confirmed both the high accuracy and the robustness of the proposed model.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2019년도 제59차 동계학술대회논문집 27권1호
• /
• pp.123-124
• /
• 2019

This paper proposes an improved deep learning method based on small data sets for animal image classification. Firstly, we use a CNN to build a training model for small data sets, and use data augmentation to expand the data samples of the training set. Secondly, using the pre-trained network on large-scale datasets, such as VGG16, the bottleneck features in the small dataset are extracted and to be stored in two NumPy files as new training datasets and test datasets. Finally, training a fully connected network with the new datasets. In this paper, we use Kaggle famous Dogs vs Cats dataset as the experimental dataset, which is a two-category classification dataset.

• 한국전자통신학회논문지
• /
• 제16권1호
• /
• pp.75-80
• /
• 2021

인공 지능은 놀라운 혜택을 제공하는 우리 삶의 중요한 부분이 되고 있다. 이와 관련하여 얼굴 표정 인식은 최근 수십 년 동안 컴퓨터 비전 연구자들 사이에서 뜨거운 주제 중 하나였다. 저해상도 이미지의 작은 데이터 세트를 분류하려면 새로운 소규모 심층 합성곱 신경망 모델을 개발해야 한다. 이를 위해 소규모 데이터 세트에 적합한 방법을 제안한다. 이 모델은 기존 심층 합성곱 신경망 모델에 비해 총 학습 가능 가중치 측면에서 메모리의 일부만 사용하지만 FER2013 및 FERPlus 데이터 세트에서 매우 유사한 결과를 보여준다.

• Genomics & Informatics
• /
• 제10권1호
• /
• pp.23-32
• /
• 2012

There are lots of studies attempting to identify the expression changes in oral squamous cell carcinoma. Most studies include insufficient samples to apply statistical methods for detecting significant gene sets. This study combined two small microarray datasets from a public database and identified significant genes associated with the progress of oral squamous cell carcinoma. There were different expression scales between the two datasets, even though these datasets were generated under the same platforms - Affymetrix U133A gene chips. We discretized gene expressions of the two datasets by adjusting the differences between the datasets for detecting the more reliable information. From the combination of the two datasets, we detected 51 significant genes that were upregulated in oral squamous cell carcinoma. Most of them were published in previous studies as cancer-related genes. From these selected genes, significant genetic pathways associated with expression changes were identified. By combining several datasets from the public database, sufficient samples can be obtained for detecting reliable information. Most of the selected genes were known as cancer-related genes, including oral squamous cell carcinoma. Several unknown genes can be biologically evaluated in further studies.

• Journal of electromagnetic engineering and science
• /
• 제18권2호
• /
• pp.136-140
• /
• 2018

We propose an improved semi-empirical scattering model for X-band radar backscattering from rough sea surfaces. This new model has a wider validity range of wind speeds than does the existing semi-empirical sea spectrum (SESS) model. First, we retrieved the small-roughness parameters from the sea surfaces, which were numerically generated using the Pierson-Moskowitz spectrum and measurement datasets for various wind speeds. Then, we computed the backscattering coefficients of the small-roughness surfaces for various wind speeds using the integral equation method model. Finally, the large-roughness characteristics were taken into account by integrating the small-roughness backscattering coefficients multiplying them with the surface slope probability density function for all possible surface slopes. The new model includes a wind speed range below 3.46 m/s, which was not covered by the existing SESS model. The accuracy of the new model was verified with two measurement datasets for various wind speeds from 0.5 m/s to 14 m/s.

• 한국음향학회지
• /
• 제42권4호
• /
• pp.329-344
• /
• 2023

본 논문에서는 소량 및 불균형 능동소나 데이터세트에 적용된 다양한 딥러닝 기반 표적식별기의 일반화 성능을 종합적으로 분석하였다. 서로 다른 시간과 해역에서 수집된 능동소나 실험 데이터를 이용하여 두 가지 능동소나 데이터세트를 생성하였다. 데이터세트의 각 샘플은 탐지 처리 이후 탐지된 오디오 신호로부터 추출된 시간-주파수 영역 이미지이다. 표적식별기의 신경망 모델은 다양한 구조를 가지는 22개의 Convolutional Neural Networks(CNN) 모델을 사용하였다. 실험에서 두 가지 데이터세트는 학습/검증 데이터세트와 테스트 데이터세트로 번갈아 가며 사용되었으며, 표적식별기 출력의 변동성을 계산하기 위해 학습/검증/테스트를 10번 반복하고 표적식별 성능을 분석하였다. 이때 학습을 위한 초매개변수는 베이지안 최적화를 이용하여 최적화하였다. 실험 결과 본 논문에서 설계한 얕은 층을 가지는 CNN 모델이 대부분의 깊은 층을 가지는 CNN 모델보다 견실하면서 우수한 일반화 성능을 가지는 것을 확인하였다. 본 논문은 향후 딥러닝 기반 능동소나 표적식별 연구에 대한 방향성을 설정할 때 유용하게 사용될 수 있다.

• 대한원격탐사학회지
• /
• 제39권1호
• /
• pp.87-97
• /
• 2023

Unmanned aerial vehicles (UAVs) can capture high-resolution imagery from a variety of viewing angles and altitudes; they are generally limited to collecting images of small scenes from larger regions. To improve the utility of UAV-appropriated datasetsfor use with deep learning applications, multiple datasets created from variousregions under different conditions are needed. To demonstrate a powerful new method for integrating heterogeneous UAV datasets, this paper applies a combined segmentation network (CSN) to share UAVid and semantic drone dataset encoding blocks to learn their general features, whereas its decoding blocks are trained separately on each dataset. Experimental results show that our CSN improves the accuracy of specific classes (e.g., cars), which currently comprise a low ratio in both datasets. From this result, it is expected that the range of UAV dataset utilization will increase.

• Genomics & Informatics
• /
• 제9권4호
• /
• pp.173-180
• /
• 2011

Recent trends in generating multiple, large-scale datasets provide new challenges to manipulating the relationship of different types of components, such as gene expression and drug response data. Integrative analysis of compound response and gene expression datasets generates an opportunity to capture the possible mechanism of compounds by using signature genes on diverse types of cancer cell lines. Here, we integrated datasets of compound response and gene expression profiles on NCI60 cell lines and constructed a network, revealing the relationship for 801 compounds and 341 gene probes. As examples, obtusol, which shows an exclusive sensitivity on a small number of colon cell lines, is related to a set of gene probes that have unique overexpression in colon cell lines. We also found that the SLC7A11 gene, a direct target of miR-26b, might be a key element in understanding the action of many diverse classes of anticancer compounds. We demonstrated that this network might be useful for studying the mechanisms of varied compound response on diverse cancer cell lines.

• ETRI Journal
• /
• 제41권3호
• /
• pp.358-370
• /
• 2019

The main aim of this study is to select the optimal set of genes from microarray cancer datasets that contribute to the prediction of specific cancer types. This study proposes the enhancement of the feature selection filter algorithm based on Joe's normalized mutual information and its use for gene selection. The proposed algorithm is implemented and evaluated on seven benchmark microarray cancer datasets, namely, central nervous system, leukemia (binary), leukemia (3 class), leukemia (4 class), lymphoma, mixed lineage leukemia, and small round blue cell tumor, using five well-known classifiers, including the naive Bayes, radial basis function network, instance-based classifier, decision-based table, and decision tree. An average increase in the prediction accuracy of 5.1% is observed on all seven datasets averaged over all five classifiers. The average reduction in training time is 2.86 seconds. The performance of the proposed method is also compared with those of three other popular mutual information-based feature selection filters, namely, information gain, gain ratio, and symmetric uncertainty. The results are impressive when all five classifiers are used on all the datasets.

• 대한원격탐사학회지
• /
• 제39권4호
• /
• pp.459-466
• /
• 2023

In the last decade, artificial intelligence's dramatic advancement with the development of various deep learning techniques has significantly contributed to remote sensing fields and satellite image applications. Among many prominent areas, super-resolution research has seen substantial growth with the release of several benchmark datasets and the rise of generative adversarial network-based studies. However, most previously published remote sensing benchmark datasets represent spatial resolution within approximately 10 meters, imposing limitations when directly applying for super-resolution of small objects with cm unit spatial resolution. Furthermore, if the dataset lacks a global spatial distribution and is specialized in particular land covers, the consequent lack of feature diversity can directly impact the quantitative performance and prevent the formation of robust foundation models. To overcome these issues, this paper proposes a method to generate benchmark datasets by simulating the modulation transfer functions of the sensor. The proposed approach leverages the simulation method with a solid theoretical foundation, notably recognized in image fusion. Additionally, the generated benchmark dataset is applied to state-of-the-art super-resolution base models for quantitative and visual analysis and discusses the shortcomings of the existing datasets. Through these efforts, we anticipate that the proposed benchmark dataset will facilitate various super-resolution research shortly in Korea.