• Korean Journal of Remote Sensing
• /
• v.38 no.1
• /
• pp.103-110
• /
• 2022

Since satellite images generally include clouds in the atmosphere, it is essential to detect or mask clouds before satellite image processing. Clouds were detected using physical characteristics of clouds in previous research. Cloud detection methods using deep learning techniques such as CNN or the modified U-Net in image segmentation field have been studied recently. Since image segmentation is the process of assigning a label to every pixel in an image, precise pixel-based dataset is required for cloud detection. Obtaining accurate training datasets is more important than a network configuration in image segmentation for cloud detection. Existing deep learning techniques used different training datasets. And test datasets were extracted from intra-dataset which were acquired by same sensor and procedure as training dataset. Different datasets make it difficult to determine which network shows a better overall performance. To verify the effectiveness of the cloud detection network such as Cloud-Net, two types of networks were trained using the cloud dataset from KOMPSAT-3 images provided by the AIHUB site and the L8-Cloud dataset from Landsat8 images which was publicly opened by a Cloud-Net author. Test data from intra-dataset of KOMPSAT-3 cloud dataset were used for validating the network. The simulation results show that the network trained with KOMPSAT-3 cloud dataset shows good performance on the network trained with L8-Cloud dataset. Because Landsat8 and KOMPSAT-3 satellite images have different GSDs, making it difficult to achieve good results from cross-sensor validation. The network could be superior for intra-dataset, but it could be inferior for cross-sensor data. It is necessary to study techniques that show good results in cross-senor validation dataset in the future.

• Land and Housing Review
• /
• v.15 no.4
• /
• pp.1-9
• /
• 2024

Visual scene understanding AI, a pivotal factor for digital transformation and robotic automation in construction, has primarily been researched under the hypothesis that the more training images, the higher the model performance. Alternatively, one can hypothesize that prioritizing activity-specific training images tailored to each construction phase would be more critical than merely enlarging the size of the dataset. This approach is particularly vital in dynamic construction environments where visual characteristics undergo significant changes across the construction phases, from earthmoving, foundation, and superstructure to finishing activities. In this background, we investigate the effects of a training image dataset's size and specificity on visual scene understanding AI in construction. We build an all-in-one, universal training image dataset as well as an activity-specific dataset, varying the number of training images. We then train vision-based worker detection models using each dataset and assess their performance in activity-specific, dynamic test environments. We analyze the optimal performance achieved in each test environment and how the model's performance varies depending on the dataset's size over the entire test phase. Our findings will help scientifically validate the dual hypotheses and lay a solid foundation for building and updating a training image dataset when developing a visual scene understanding AI model in dynamic construction sites.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.01a
• /
• 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.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.27 no.4
• /
• pp.311-323
• /
• 2023

This paper introduces the industrial problem, the solution, and the results of the research conducted with Define Inc. The client company wanted to improve the performance of an object detection model on the floor plan dataset. To solve the problem, we analyzed the operational principles, advantages, and disadvantages of the existing object detection model, identified the characteristics of the floor plan dataset, and proposed to use of YOLO v5 as an appropriate object detection model for training the dataset. We compared the performance of the existing model and the proposed model using mAP@60, and verified the object detection results with real test data, and found that the performance increase of mAP@60 was 0.08 higher with a 25% shorter inference time. We also found that the training time of the proposed YOLO v5 was 71% shorter than the existing model because it has a simpler structure. In this paper, we have shown that the object detection model for the floor plan dataset can achieve better performance while reducing the training time. We expect that it will be useful for solving other industrial problems related to object detection in the future. We also believe that this result can be extended to study object recognition in 3D floor plan dataset.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.1
• /
• pp.99-107
• /
• 2023

Recently, the rate of death and safety accidents at construction sites is the highest among all kinds of industries. In order to apply artificial intelligence technology to construction sites, it is essential to secure a dataset which can be used as a basic training data. In this paper, a number of image data were collected through actual construction site, for which major construction equipment objects mainly operated in civil engineering sites were defined. The optimal training dataset construction was completed by annotation process of about 90,000 image dataset. Reliability of the dataset was verified with the mAP of over 90 % in use of YOLO, a representative model in the field of object detection. The construction equipment training dataset built in this study has been released which is currently available on the public data portal of the Ministry of Public Administration and Security. This dataset is expected to be freely used for any application of object detection technology on construction sites especially in the field of construction safety in the future.

• Annual Conference of KIPS
• /
• 2017.04a
• /
• pp.787-790
• /
• 2017

Data quality is the main issue in the classification problems; generally, the presence of noisy instances in the training dataset will not lead to robust classification performance. Such instances may cause the generated decision tree to suffer from over-fitting and its accuracy may decrease. Decision trees are useful, efficient, and commonly used for solving various real world classification problems in data mining. In this paper, we introduce a preprocessing technique to improve the classification accuracy rates of the C4.5 decision tree algorithm. In the proposed preprocessing method, we applied the naive Bayes classifier to remove the noisy instances from the training dataset. We applied our proposed method to a real e-commerce sales dataset to test the performance of the proposed algorithm against the existing C4.5 decision tree classifier. As the experimental results, the proposed method improved the classification accuracy by 8.5% and 14.32% using training dataset and 10-fold crossvalidation, respectively.

• Korean Journal of Remote Sensing
• /
• v.38 no.4
• /
• pp.375-386
• /
• 2022

The city of Khartoum, the capital of Sudan, was heavily damaged by the flood of the Nile in 2020. Classification using satellite images can define the damaged area and help emergency response. As Synthetic Aperture Radar (SAR) uses microwave that can penetrate cloud, it is suitable to use in the flood study. In this study, Random Forest classifier, one of the supervised classification algorithms, was applied to the flood event in Khartoum with various sizes of the training dataset and number of images using Sentinel-1 SAR. To create a training dataset, we used unsupervised classification and visual inspection. Firstly, Random Forest was performed by reducing the size of each class of the training dataset, but no notable difference was found. Next, we performed Random Forest with various number of images. Accuracy became better as the number of images in creased, but converged to a maximum value when the dataset covers the duration from flood to the completion of drainage.

• Journal of the Korean Society for Library and Information Science
• /
• v.54 no.2
• /
• pp.419-434
• /
• 2020

The purpose of the study is to design and propose metadata standard for building AI training dataset of landmark images. In order to achieve the purpose, we first examined and analyzed the state of art of the types of image retrieval systems and their indexing methods, comprehensively. We then investigated open training dataset and machine learning tools for image object recognition. Sequentially, we selected metadata elements optimized for the AI training dataset of landmark images and defined the input data for each element. We then concluded the study with implications and suggestions for the development of application services using the results of the study.

• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.293-298
• /
• 2023

The military is facing a continuous decrease in personnel, and in order to cope with potential accidents and challenges in operations, efforts are being made to reduce the direct involvement of personnel by utilizing the latest technologies. Recently, the use of various sensors related to Manned-Unmanned Teaming and artificial intelligence technologies has gained attention, emphasizing the need for flexible utilization methods. In this paper, we propose four dataset construction methods that can be used for effective training of robots that can be deployed in military operations, utilizing not only RGB image data but also data acquired from IR image sensors. Since there is no publicly available dataset that combines RGB and IR image data, we directly acquired the dataset within buildings. The input values were constructed by combining RGB and IR image sensor data, taking into account the field of view, resolution, and channel values of both sensors. We compared the proposed method with conventional RGB image data classification training using the same learning model. By employing the proposed image data fusion method, we observed improved stability in training loss and approximately 3% higher accuracy.

• Journal of the Semiconductor & Display Technology
• /
• v.23 no.3
• /
• pp.135-139
• /
• 2024

Public datasets, which are freely available and often labeled, play a crucial role in training object detection models in computer vision. While public datasets are effective for developing general object detection models, they may not be ideal for specialized tasks. For specific object detection needs, it is more beneficial to create and use a dataset tailored to the target object. This paper proposes a method for extracting a target-specific dataset from public datasets to develop object detection models with superior performance for the target object. This approach not only improves detection accuracy, but also reduces training data requirements and complexity. We evaluate the performance of the proposed method using the latest object detection model YOLOv10.