• Journal of the Korean Society for Precision Engineering
• /
• v.30 no.9
• /
• pp.901-908
• /
• 2013

The introduction of industrial intelligent robot using vision sensor has been interested in automated factory. 2D and 3D vision sensors have used to recognize object and to estimate object pose, which is for packaging parts onto a complete whole. But it is not trivial task due to illumination and various types of objects. Object image has distorted due to illumination that has caused low reliability in recognition. In this paper, recognition method of complex shape object has been proposed. An accurate object region has detected from combined binary image, which has achieved using DoG filter and local adaptive binarization. The object has recognized using neural network, which is trained with sub-divided object class according to object type and rotation angle. Predefined shape model of object and maximal slope have used to estimate the pose of object. The performance has evaluated on ETRI database and recognition rate of 96% has obtained.

• Fashion, Industry and Education
• /
• v.14 no.1
• /
• pp.59-68
• /
• 2016

This paper identified the source of inspiration to cooperatively design a fashion collection from US undergraduate clothing design students and addressed how to implement team-based learning strategy to conceptual clothing design in class. Data was collected from the total of 51 students in a senior clothing design course at a large 4-year university in the US. The assigned project for this class was to develop a group collection under a same theme. Each student worked with his/her team member(s) to create an outfit and the entire class worked as a group to create a cohesive collection. The study showed that the sources of inspiration for the themes/concepts came from 11categories: historic era/old Hollywood glamour, shape/line/structure/architectural, fairy tales movies, nature/abstract, circus/mysterious, occasion/place, object, designer/artist, futuristic, culture, and various movies. To implement cooperative learning strategies in the clothing design class, a total of five class presentation/discussion sessions were held for theme/concept decision, fabric decision, design decision, test garment evaluation and design modification, and final products. Throughout the design process, team-based learning strategy promoted students' engagement and participation and inspired their critical thinking skills for making decisions within a team.

• Bulletin of the Korean Mathematical Society
• /
• v.41 no.1
• /
• pp.53-71
• /
• 2004

A certain subclass $T_{\Omega}(n,\;p,\;\lambda,\;\alpha)$ of starlike functions in the unit disk is introduced. The object of the present paper is to derive several interesting properties of functions belonging to the class $T_{\Omega}(n,\;p,\;\lambda,\;\alpha)$. Coefficient inequalities, distortion theorems and closure theorems of functions belonging to the class $T_{\Omega}(n,\;p,\;\lambda,\;\alpha)$ are determined. Also we obtain radii of convexity for the class $T_{\Omega}(n,\;p,\;\lambda,\;\alpha)$. Furthermore, integral operators and modified Hadamard products of several functions belonging to the class $T_{\Omega}(n,\;p,\;\lambda,\;\alpha)$ are studied here.

• Journal of the Korea Society of Computer and Information
• /
• v.9 no.2
• /
• pp.49-56
• /
• 2004

In this paper, we design edge class that can processed digital image effectively, edge is a important information including the point of shape information for a object detection or recognition in the digital image. Therefore, it is of very importance, which managed effectively the edge and can use a variety availability in digital image Processing, after edge detection. The environment using the existing digital image processing system has limits of use and speed. In this paper, we design edge class that can managed detected edges and it analyzes existing methods by edge detection algorithm.

• Special Issue of the Society of Naval Architects of Korea
• /
• 2006.09a
• /
• pp.40-45
• /
• 2006

For ships of ice class, finish Maritime Administration(FMA) requires brackets on intersections between longitudinal frames and the web frames within the ice-strengthened area. The main object of this paper is to verify ultimate load carrying capacity of longitudinal frame without brackets of engine room region of 74,100 DWT Product Oil Tanker. Comparative approach between proposed structures from builder (the proposed structure) and structures satisfying the Finnish-Swedish ice class rules (the rule structure) is used for the analysis.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.12
• /
• pp.6009-6027
• /
• 2019

Support vector machine (SVM) classifiers have been widely used for object detection. These methods usually locate the object by finding the region with maximal score in an image. With bag-of-features representation, the SVM score of an image region can be written as the sum of its inside feature-weights. As a result, the searching process can be executed efficiently by using strategies such as branch-and-bound. However, the feature-weight derived by optimizing region classification cannot really reveal the category knowledge of a feature-point, which could cause bad localization. In this paper, we represent a region in an image by a collection of local feature-points and determine the object by the region with the maximum posterior probability of belonging to the object class. Based on the Bayes' theorem and Naive-Bayes assumptions, the posterior probability is reformulated as the sum of feature-scores. The feature-score is manifested in the form of the logarithm of a probability ratio. Instead of estimating the numerator and denominator probabilities separately, we readily employ the density ratio estimation techniques directly, and overcome the above limitation. Experiments on a car dataset and PASCAL VOC 2007 dataset validated the effectiveness of our method compared to the baselines. In addition, the performance can be further improved by taking advantage of the recently developed deep convolutional neural network features.

• The Journal of Korea Robotics Society
• /
• v.14 no.2
• /
• pp.139-149
• /
• 2019

In this paper, we present auto-annotation tool and synthetic dataset using 3D CAD model for deep learning based object detection. To be used as training data for deep learning methods, class, segmentation, bounding-box, contour, and pose annotations of the object are needed. We propose an automated annotation tool and synthetic image generation. Our resulting synthetic dataset reflects occlusion between objects and applicable for both underwater and in-air environments. To verify our synthetic dataset, we use MASK R-CNN as a state-of-the-art method among object detection model using deep learning. For experiment, we make the experimental environment reflecting the actual underwater environment. We show that object detection model trained via our dataset show significantly accurate results and robustness for the underwater environment. Lastly, we verify that our synthetic dataset is suitable for deep learning model for the underwater environments.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.12
• /
• pp.1588-1595
• /
• 2019

This paper presents the technique on structuring information of the diagram repository for UML diagrams. Because object interactions are the body of object-oriented programming, this paper handles especially the sequence diagrams and class diagrams among UML diagrams. Based on class diagrams, sequence diagrams represent the procedure of object interactions in run-time and then the corresponding codes are generated from the contents of those sequence diagrams. To do this work, this paper presents a method to construct the information repository for generating code from the contents of sequence diagrams. This paper classifies the five message types of sequence diagrams and then extracts the needed information including items and values on the corresponding message types for constructing message repositories. Because sequence diagram is composed of messages included, the final repository is constructed by collecting each of structured repositories on messages sequentially.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.12
• /
• pp.77-84
• /
• 2022

In this paper, we propose a Label Reassignment Strategy to improve the performance of an object detection algorithm. Our approach involves two stages: an inference stage and an assignment stage. In the inference stage, we perform multi-scale inference with predefined scale sizes on a trained model and re-infer masked images to obtain robust classification results. In the assignment stage, we calculate the IoU between bounding boxes to remove duplicates. We also check box and class occurrence between the detection result and annotation label to re-assign the dominant class type. We trained the YOLOX-L model with the re-annotated dataset to validate our strategy. The model achieved a 3.9% improvement in mAP and 3x better performance on AP_S compared to the model trained with the original dataset. Our results demonstrate that the proposed Label Reassignment Strategy can effectively improve the performance of an object detection model.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.3
• /
• pp.571-580
• /
• 2022

Marine deposited waste is a major cause of problems such as a lot of damage and an increase in the estimated amount of garbage due to abandoned fishing grounds caused by ghost fishing. In this paper, we implement a real-time marine deposited waste detection artificial intelligence system to understand the actual conditions of waste fishing gear usage, distribution, loss, and recovery, and study methods for performance improvement. The system was implemented using the yolov5 model, which is an excellent performance model for real-time object detection, and the 'data screening process' and 'class segmentation' method of learning data were applied as performance improvement methods. In conclusion, the object detection results of datasets that do screen unnecessary data or do not subdivide similar items according to characteristics and uses are better than the object recognition results of unscreened datasets and datasets in which classes are subdivided.