• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.44-55
• /
• 2022

In order to automatically provide information on fruits in agricultural related broadcasting contents, instance image segmentation of target fruits is required. In addition, the information on the 3D pose of the corresponding fruit may be meaningfully used. This paper represents research that provides information about tomatoes in video content. A large amount of data is required to learn the instance segmentation, but it is difficult to obtain sufficient training data. Therefore, the training data is generated through a data augmentation technique based on a small amount of real images. Compared to the result using only the real images, it is shown that the detection performance is improved as a result of learning through the synthesized image created by separating the foreground and background. As a result of learning augmented images using images created using conventional image pre-processing techniques, it was shown that higher performance was obtained than synthetic images in which foreground and background were separated. To estimate the pose from the result of object detection, a point cloud was obtained using an RGB-D camera. Then, cylinder fitting based on least square minimization is performed, and the tomato pose is estimated through the axial direction of the cylinder. We show that the results of detection, instance image segmentation, and cylinder fitting of a target object effectively through various experiments.

• Annual Conference of KIPS
• /
• 2005.11a
• /
• pp.477-480
• /
• 2005

메모리 기반 추론 기법에서 기억공간의 효율적 사용과 분류 시간을 줄이기 위한 다양한 방법들이 연구되고 있으며, NGE(Nested Generalized Exemplar) 이론을 예로 들 수 있다. 본 논문에서는 학습 패턴 집합으로부터 대표패턴을 생성하는 RPA(Recursive Partition Averaging) 기법과 점진적으로 대표패턴을 추출하는 IRPA(Incremental RPA) 기법을 제안한다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.38 no.6
• /
• pp.635-644
• /
• 2020

Object recognition, detection and instance segmentation based on DL (Deep Learning) have being used in various practices, and mainly optical images are used as training data for DL models. The major objective of this paper is object segmentation and building detection by utilizing multimodal datasets as well as optical images for training Detectron2 model that is one of the improved R-CNN (Region-based Convolutional Neural Network). For the implementation, infrared aerial images, LiDAR data, and edges from the images, and Haralick features, that are representing statistical texture information, from LiDAR (Light Detection And Ranging) data were generated. The performance of the DL models depends on not only on the amount and characteristics of the training data, but also on the fusion method especially for the multimodal data. The results of segmenting objects and detecting buildings by applying hybrid fusion - which is a mixed method of early fusion and late fusion - results in a 32.65% improvement in building detection rate compared to training by optical image only. The experiments demonstrated complementary effect of the training multimodal data having unique characteristics and fusion strategy.

• Convergence Security Journal
• /
• v.22 no.2
• /
• pp.21-26
• /
• 2022

Wearing face masks is an effective measure to prevent COVID-19 infection. Infrared thermal image based temperature measurement and identity recognition system has been widely used in many large enterprises and universities in China, so it is totally necessary to research the face mask detection of thermal infrared imaging. Recently introduced MTCNN (Multi-task Cascaded Convolutional Networks)presents a conceptually simple, flexible, general framework for instance segmentation of objects. In this paper, we propose an algorithm for efficiently searching objects of images, while creating a segmentation of heat generation part for an instance which is a heating element in a heat sensed image acquired from a thermal infrared camera. This method called a mask MTCNN is an algorithm that extends MTCNN by adding a branch for predicting an object mask in parallel with an existing branch for recognition of a bounding box. It is easy to generalize the R-CNN to other tasks. In this paper, we proposed an infrared image detection algorithm based on R-CNN and detect heating elements which can not be distinguished by RGB images.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.207-220
• /
• 2022

Structural health monitoring (SHM) plays a vital role in the maintenance and operation of constructions. In recent years, autonomous inspection has received considerable attention because conventional monitoring methods are inefficient and expensive to some extent. To develop autonomous inspection, a potential approach of crack identification is needed to locate defects. Therefore, this study exploits two deep learning-based segmentation models, DeepLabv3+ and Mask R-CNN, for crack segmentation because these two segmentation models can outperform other similar models on public datasets. Additionally, impacts of label quality on model performance are explored to obtain an empirical guideline on the preparation of image datasets. The influence of image cropping and label refining are also investigated, and different strategies are applied to the dataset, resulting in six alternated datasets. By conducting experiments with these datasets, the highest mean Intersection-over-Union (mIoU), 75%, is achieved by Mask R-CNN. The rise in the percentage of annotations by image cropping improves model performance while the label refining has opposite effects on the two models. As the label refining results in fewer error annotations of cracks, this modification enhances the performance of DeepLabv3+. Instead, the performance of Mask R-CNN decreases because fragmented annotations may mistake an instance as multiple instances. To sum up, both DeepLabv3+ and Mask R-CNN are capable of crack identification, and an empirical guideline on the data preparation is presented to strengthen identification successfulness via image cropping and label refining.

• Journal of Korea Multimedia Society
• /
• v.6 no.5
• /
• pp.824-830
• /
• 2003

RPA (Recursive Partition Averaging) method was proposed in order to improve the storage requirement and classification rate of the Memory Based Reasoning. That algorithm worked well in many areas, however, the major drawbacks of RPA are it's pattern averaging mechanism. We propose an adaptive OHC algorithm which uses the FPD(Feature-based Population Densimeter) to increase the classification rate of RPA. The proposed algorithm required only approximately 40% of memory space that is needed in k-NN classifier, and showed a superior classification performance to the RPA. Also, by reducing the number of stored patterns, it showed a excellent results in terms of classification when we compare it to the k-NN.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.371-375
• /
• 2008

예제 기반 학습(instance-based learning) 방법 중 하나인 k-최근접 이웃(k-nearest reighbor, k-NN) 학습은 간단하고 예측 정확도가 비교적 높아 분류 및 회귀 문제 해결을 위한 기반 방법론으로 널리 적용되고 있다. k-NN 학습을 위한 알고리즘은 기본적으로 유클리드 거리 혹은 2-놈(norm)에 기반하여 학습예제들 사이의 거리를 계산한다. 본 논문에서는 유클리드 거리를 일반화한 개념인 p-놈의 사용이 k-NN 학습의 성능에 어떠한 영향을 미치는지 연구하였다. 구체적으로 합성데이터와 다수의 기계학습 벤치마크 문제 및 실제 데이터에 다양한 p-놈을 적용하여 그 일반화 성능을 경험적으로 조사하였다. 실험 결과, 데이터에 잡음이 많이 존재하거나 문제가 어려운 경우에 p의 값을 작게 하는 것이 성능을 향상시킬 수 있었다.

• Smart Media Journal
• /
• v.12 no.3
• /
• pp.112-119
• /
• 2023

Shrimp farming has been becoming a new source of income for fishermen in South Korea. It is often necessary for fishers to measure the size of the shrimp for the purpose to understand the growth rate of the shrimp and to determine the amount of food put into the breeding pond. Traditional methods rely on humans, which has huge time and labor costs. This paper proposes a deep learning-based method for calculating the size of shrimps automatically. Firstly, we use fine-tuning techniques to update the Mask RCNN model with our farm data, enabling it to segment shrimps and generate shrimp masks. We then use skeletonizing method and maximum inscribed circle to calculate the length and width of shrimp, respectively. Our method is simple yet effective, and most importantly, it requires a small hardware resource and is easy to deploy to shrimp farms.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.2 s.34
• /
• pp.31-38
• /
• 2005

The web is expected to cause the lot of its utilizing as a means of computer instructed learning and recently the applying instance on the web is more increasing in the education. From educational perspectives. the web-based instruction is much superior to any other medium in the view of interaction and greater to the amount of transmitted information. The various learning contents of WBI program can make students feel more excited and interested in learning activities. Also the creative talent and application abilities of the learner are able to be developed by mixing various sorts of multimedia materials up such as moving pictures. graphics and sounds. In this study, a WBI courseware learning program for the problem solving was designed and Proposed on the base of the theory of constructivism for the subject of social science in the 6th grade of elementary school. The experimental results showed that the learning accomplishment of an experimental class was much better than that of an existent class.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.6 no.2
• /
• pp.13-21
• /
• 1996

The necessity and usefulness of higher-order neural networks have been well-known since early days of neurocomputing. However the explosive number of terms has hampered the design and training of such networks. In this paper we present an evolutionary learning method for efficiently constructing problem-specific higher-order neural models. The crux of the method is the neural tree representation employing both sigma and pi units, in combination with the use of an MDL-based fitness function for learning minimal models. We provide experimental results in classification and prediction problems which demonstrate the effectiveness of the method. I. Introduction topology employs one hidden layer with full connectivity between neighboring layers. This structure has One of the most popular neural network models been very successful for many applications. However, used for supervised learning applications has been the they have some weaknesses. For instance, the fully mutilayer feedforward network. A commonly adopted connected structure is not necessarily a good topology unless the task contains a good predictor for the full *d*dWs %BH%W* input space.