• Land and Housing Review
• /
• v.14 no.4
• /
• pp.95-102
• /
• 2023

Numerous investigations have explored the integration of machine leaning algorithms with high-resolution drone image for object detection in urban settings. However, a prevalent limitation in vehicle extraction studies involves the reliance on bounding boxes rather than instance segmentation. This limitation hinders the precise determination of vehicle direction and exact boundaries. Instance segmentation, while providing detailed object boundaries, necessitates labour intensive labelling for individual objects, prompting the need for research on automating unsupervised instance segmentation in vehicle extraction. In this study, a novel approach was proposed for vehicle extraction utilizing unsupervised SVM classification applied to vehicle bounding boxes in drone images. The method aims to address the challenges associated with bounding box-based approaches and provide a more accurate representation of vehicle boundaries. The study showed promising results, demonstrating an 89% accuracy in vehicle extraction. Notably, the proposed technique proved effective even when dealing with significant variations in spectral characteristics within the vehicles. This research contributes to advancing the field by offering a viable solution for automatic and unsupervised instance segmentation in the context of vehicle extraction from image.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.9
• /
• pp.2299-2308
• /
• 1997

While most machine learning research has been primarily concerned with the development of systems that implement one type of learning strategy, we use a multistrategy approach which integrates rule induction learning and instance-based learning, and show how this marriage allows for overall better performance. In the rule induction learning phase, we derive an entropy function, based on Hellinger divergence, which can measure the amount of information each inductive rule contains, and show how well the Hellinger divergence measures the importance of each rule. We also propose some heuristics to reduce the computational complexity by analyzing the characteristics of the Hellinger measure. In the instance-based learning phase, we improve the current instance-based learning method in a number of ways. The system has been implemented and tested on a number of well-known machine learning data sets. The performance of the system has been compared with that of other classification learning technique.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2020.11a
• /
• pp.55-56
• /
• 2020

The construction industry has the highest occupational fatality and injury rates related to accidents of any industry. Accordingly, safety managers closely monitor to prevent accidents in real-time by installing surveillance cameras at construction sites. However, due to human cognitive ability limitations, it is impossible to monitor many videos simultaneously, and the fatigue of the person monitoring surveillance cameras is also very high. Thus, to help safety managers monitor work and reduce the occupational accident rate, a study on object recognition in construction sites was conducted through surveillance cameras. In this study, we applied to the instance segmentation to identify the classification and location of objects and extract the size and shape of objects in construction sites. This research considers ways in which deep learning-based computer vision technology can be applied to safety management on a construction site.

• Journal of Information Technology Applications and Management
• /
• v.26 no.2
• /
• pp.111-123
• /
• 2019

In this paper, software activity, software activity instance, and the quality of the activity instance are defined as the 'activity which is performed on the software product by a person or a group of persons,' the 'distinctive and individual performance of software activity,' and the 'performer's evaluation on how good or bad his/her own activity instance is,' respectively. The representative values of the instance quality population associated with a product and its sub-population are defined as the (software) activity quality and activity quality characteristic of the product, respectively. The activity quality model in this paper classifies activity quality characteristics according to the classification hierarchy of software activity by the goal. In the model, a quality characteristic can have two types of sub-characteristics : Special sub-characteristic and component sub-characteristic, where the former is its super-characteristic too simultaneously and the latter is not its super-characteristic but a part of its super-characteristic. The activity quality model is parsimonious, coherent, and easy to understand and use. The activity quality model can serve as a corner stone on which a software quality body of knowledge, which constituted with a set of models parsimonious, coherent, and easy to understand and use and the theories explaining the cause-and-relationships among the models, can be built. The body of knowledge can be called the (grand) activity-centric quality model of software.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.363-365
• /
• 2003

This paper presents an object oriented approach(OOA) for classification of multi-channel and multi-polarization NASA/JPL POLSAR images. Some test results in Taiwan are also given and analyzed. It is concluded that this approach can utilize as more information of both low- and high-levels involved in all images as possible for image classification and thus provides a better classification accuracy. For instance, the OOA has a better overall classification accuracy(98.27%) than the nearest-neighbor classifier(91.31%) and minimum-distance classifier(80.52%).

• Journal of Information Processing Systems
• /
• v.7 no.4
• /
• pp.717-732
• /
• 2011

Classification in biomedical applications is an important task that predicts or classifies an outcome based on a given set of input variables such as diagnostic tests or the symptoms of a patient. Traditionally the classification algorithms would have to digest a stationary set of historical data in order to train up a decision-tree model and the learned model could then be used for testing new samples. However, a new breed of classification called stream-based classification can handle continuous data streams, which are ever evolving, unbound, and unstructured, for instance--biosignal live feeds. These emerging algorithms can potentially be used for real-time classification over biosignal data streams like EEG and ECG, etc. This paper presents a pioneer effort that studies the feasibility of classification algorithms for analyzing biosignals in the forms of infinite data streams. First, a performance comparison is made between traditional and stream-based classification. The results show that accuracy declines intermittently for traditional classification due to the requirement of model re-learning as new data arrives. Second, we show by a simulation that biosignal data streams can be processed with a satisfactory level of performance in terms of accuracy, memory requirement, and speed, by using a collection of stream-mining algorithms called Optimized Very Fast Decision Trees. The algorithms can effectively serve as a corner-stone technology for real-time classification in future biomedical applications.

• Journal of Korea Society of Industrial Information Systems
• /
• v.10 no.1
• /
• pp.28-37
• /
• 2005

In our previous works, we suggested a method that extracts representative keywords from a few positive documents and assigns weights to them. To show the usefulness of the method, in this paper, we evaluate the performance of a famous classification algorithm called GIS(Generalized Instance Set) when it is combined with our method. In GIS algorithm, generalized instances are built from learning documents by a generalization function and then the K-NN algorithm is applied to them. Here, our method is used as a generalization function. For comparative works, Rocchio and Widrow-Hoff algorithms are also used as a generalization function. Experimental results show that our method is better than the others for the case that only positive documents are considered, but not when negative documents are considered together.

• Journal of the Korean Society of Safety
• /
• v.34 no.6
• /
• pp.50-57
• /
• 2019

In this study, various types of deep learning models that have been proposed recently are classified according to data input / output types and analyzed to find the deep learning model suitable for constructing a crack detection model. First the deep learning models are classified into image classification model, object segmentation model, object detection model, and instance segmentation model. ResNet-101, DeepLab V2, Faster R-CNN, and Mask R-CNN were selected as representative deep learning model of each type. For the comparison, ResNet-101 was implemented for all the types of deep learning model as a backbone network which serves as a main feature extractor. The four types of deep learning models were trained with 500 crack images taken from real concrete structures and collected from the Internet. The four types of deep learning models showed high accuracy above 94% during the training. Comparative evaluation was conducted using 40 images taken from real concrete structures. The performance of each type of deep learning model was measured using precision and recall. In the experimental result, Mask R-CNN, an instance segmentation deep learning model showed the highest precision and recall on crack detection. Qualitative analysis also shows that Mask R-CNN could detect crack shapes most similarly to the real crack shapes.

• Journal of Advanced Marine Engineering and Technology
• /
• v.28 no.4
• /
• pp.611-617
• /
• 2004

Pixel classification is one of basic image processing issues. The general characteristics of the pixels belonging to various classes are discussed and the radical principles of pixel classification are given. At the same time. a pixel classification scheme based on image direction measure is proposed. As a typical application instance of pixel classification, an adaptive multi-level median filter is presented. An image can be classified into two types of areas by using the direction information measure, that is. smooth area and edge area. Single direction multi-level median filter is used in smooth area. and multi-direction multi-level median filter is taken in the other type of area. What's more. an adaptive mechanism is proposed to adjust the type of the filters and the size of filter window. As a result. we get a better trade-off between preserving details and noise filtering.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.383-386
• /
• 2005

This study proposes a case based reasoning system with two dimensional reduction techniques. In this study, vertical and horizontal dimensions of the research data are reduced through hybrid feature and instance selection process using genetic algorithms. We applied the proposed model to customer classification model which utilizes customers' demographic characteristics as inputs to predict their buying behavior for the specific product. Experimental results show that the proposed technique may improve the classification accuracy and outperform various optimized models of typical CBR system.