• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.6
• /
• pp.813-826
• /
• 2022

AI-based image processing technologies in various fields have been widely studied. However, the lighter the board, the more difficult it is to reduce the weight of image processing algorithm due to a lot of computation. In this paper, we propose a method using deep learning for object recognition algorithm in lightweight embedded boards. We can determine the area using a deep neural network architecture algorithm that processes semantic segmentation with a relatively small amount of computation. After masking the area, by using more accurate deep learning algorithm we could operate object detection with improved accuracy for efficient neural network (ENet) and You Only Look Once (YOLO) toward executing object recognition in real time for lightweighted embedded boards. This research is expected to be used for autonomous driving applications, which have to be much lighter and cheaper than the existing approaches used for object recognition.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.153-166
• /
• 2022

Identifying fine cracks in steel bridge facilities is a challenging task of structural health monitoring (SHM). This study proposed an end-to-end crack image segmentation framework based on a one-step Convolutional Neural Network (CNN) for pixel-level object recognition with high accuracy. To particularly address the challenges arising from small object detection in complex background, efforts were made in loss function selection aiming at sample imbalance and module modification in order to improve the generalization ability on complicated images. Specifically, loss functions were compared among alternatives including the Binary Cross Entropy (BCE), Focal, Tversky and Dice loss, with the last three specialized for biased sample distribution. Structural modifications with dilated convolution, Spatial Pyramid Pooling (SPP) and Feature Pyramid Network (FPN) were also performed to form a new backbone termed CrackDet. Models of various loss functions and feature extraction modules were trained on crack images and tested on full-scale images collected on steel box girders. The CNN model incorporated the classic U-Net as its backbone, and Dice loss as its loss function achieved the highest mean Intersection-over-Union (mIoU) of 0.7571 on full-scale pictures. In contrast, the best performance on cropped crack images was achieved by integrating CrackDet with Dice loss at a mIoU of 0.7670.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.11
• /
• pp.1-7
• /
• 2018

In this paper, we propose soft dead pixels detection method by analysing different response at hot and cold images. Abnormal pixels are able to effect detecting a small target. It also makes confusing real target or not cause of changing target size. Almost exist abnormal pixels after image signal processing even if dead pixels are removed by dead pixel compensation are called soft dead pixels. They are showed defect in final image. So removing or compensating dead pixels are very important for detecting object. The key idea of this proposed method, detecting dead pixels, is that most of soft deads have different response characteristics between hot image and cold image. General infrared cameras do NUC to remove FPN. Working 2-reference NUC must be needed getting data, hot & cold images. The way which is proposed dead pixel detection is that we compare response, NUC gain, at each pixel about two different temperature images and find out dead pixels if the pixels exceed threshold about average gain of around pixels.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.5
• /
• pp.265-272
• /
• 2022

As the risk of lung cancer has increased, early-stage detection and treatment of cancers have received a lot of attention. Among various medical imaging approaches, computer tomography (CT) has been widely utilized to examine the size and growth rate of lung nodules. However, the process of manual examination is a time-consuming task, and it causes physical and mental fatigue for medical professionals. Recently, many computer-aided diagnostic methods have been proposed to reduce the workload of medical professionals. In recent studies, encoder-decoder architectures have shown reliable performances in medical image segmentation, and it is adopted to predict lesion candidates. However, localizing nodules in lung CT images is a challenging problem due to the extremely small sizes and unstructured shapes of nodules. To solve these problems, we utilize atrous spatial pyramid pooling (ASPP) to minimize the loss of information for a general U-Net baseline model to extract rich representations from various receptive fields. Moreover, we propose mixed-up attention mechanism of reverse, boundary and convolutional block attention module (CBAM) to improve the accuracy of segmentation small scale of various shapes. The performance of the proposed model is compared with several previous attention mechanisms on the LIDC-IDRI dataset, and experimental results demonstrate that reverse, boundary, and CBAM (RB-CBAM) are effective in the segmentation of small nodules.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.4
• /
• pp.523-529
• /
• 2021

In the era of the Fourth Industrial Revolution, technological changes are being applied in various fields. Automation digitization and data management are also in the field of bills. There are more than tens of thousands of forms of bills circulating in society and bill recognition is essential for automation, digitization and data management. Currently in order to manage various bills, OCR technology is used for character recognition. In this time, we can increase the accuracy, when firstly recognize the form of the bill and secondly recognize bills. In this paper, a logo that can be used as an index to classify the form of the bill was recognized as an object. At this time, since the size of the logo is smaller than that of the entire bill, FPN was used for Small Object Detection among deep learning technologies. As a result, it was possible to reduce resource waste and increase the accuracy of OCR recognition through the proposed algorithm.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.391-401
• /
• 2022

As deep learning-based object detection and recognition research have been developed recently, the scope of application to industry and real life is expanding. But deep learning-based systems in the construction system are still much less studied. Calculating materials in the construction system is still manual, so it is a reality that transactions of wrong volumn calculation are generated due to a lot of time required and difficulty in accurate accumulation. A fast and accurate automatic drawing recognition system is required to solve this problem. Therefore, we propose an AI-based automatic drawing recognition accumulation system that detects and recognizes steel materials in construction drawings. To accurately detect steel materials in construction drawings, we propose data augmentation techniques and spatial attention modules for improving small object detection performance based on YOLOv4. The detected steel material area is recognized by text, and the number of steel materials is integrated based on the predicted characters. Experimental results show that the proposed method increases the accuracy and precision by 1.8% and 16%, respectively, compared with the conventional YOLOv4. As for the proposed method, Precision performance was 0.938. The recall was 1. Average Precision AP_0.5 was 99.4% and AP_0.5:0.95 was 67%. Accuracy for character recognition obtained 99.9.% by configuring and learning a suitable dataset that contains fonts used in construction drawings compared to the 75.6% using the existing dataset. The average time required per image was 0.013 seconds in the detection, 0.65 seconds in character recognition, and 0.16 seconds in the accumulation, resulting in 0.84 seconds.

• Korean Journal of Computational Design and Engineering
• /
• v.18 no.5
• /
• pp.374-383
• /
• 2013

Various interaction techniques have been studied for providing the feeling of touch and improve immersion in augmented reality (AR) environments. Tangible AR interaction exploiting two types (product-type and pointer-type) of simple objects has earned great interest for cost-effective design evaluation of digital handheld products. When the sizes of markers attached to the objects are kept big to obtain better marker recognition, the pointer-type object frequently and significantly occludes the product-type object, which deteriorates natural visualization and level of immersion in an AR environment. In this paper, in order to overcome such problems, we propose tangible AR interaction using fingertip touch combined with small-sized markers. The proposed approach facilitates the use of convex polygons to recover the boundaries of AR markers which are partially occluded. It also properly enlarges the pattern area of each AR marker to reduce the sizes of AR markers without sacrificing the quality of marker detection. We empirically verified the quality of the proposed approach, and applied it in the process of design evaluation of digital products. From experimental results, we found that the approach is comparably accurate enough to be applied to the design evaluation process and tangible enough to provide a pseudo feeling of manipulating virtual products with human hands.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.3
• /
• pp.249-255
• /
• 2013

In this paper, we propose an algorithm to detect scale invariant object from IR image obtained in the sea environment. We create horizontal edge (HL), vertical edge (LH), diagonal edge (HH) of images through 2-D discrete Haar wavelet transform (DHWT) technique after noise reduction using morphology operations. Considering the sea environment, Gaussian blurring to the horizontal and vertical edge images at each level of wavelet is performed and then saliency map is generated by multiplying the blurred horizontal and vertical edges and combining into one image. Then we extract object candidate region by performing a binarization to saliency map. A small area in the object candidate region are removed to produce final result. Experiment results show the feasibility of the proposed algorithm.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.817-822
• /
• 2005

Recently tile study that exploits visual information for tile services of robot in indoor environments is active. Conventional image processing approaches are based on the pre-defined geometric models, so their performances are likely to decrease when they are applied to the uncertain and dynamic environments. For this, diverse researches to manage the uncertainty based on the knowledge for improving image recognition performance have been doing. In this paper we propose a Bayesian network modeling method for predicting the existence of target objects when they are occluded by other ones for improving the object detection performance of the service robots. The proposed method makes object relationship, so that it allows to predict the target object through observed ones. For this, we define the design method for small size Bayesian networks (primitive Bayesian netqork), and allow to integrate them following to the situations. The experiments are performed for verifying the performance of constructed model, and they shows $82.8\%$ of accuracy in 5 places.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.1
• /
• pp.1-8
• /
• 2024

Robotic arms have been widely utilized in various labor-intensive industries such as manufacturing, agriculture, and food services, contributing to increasing productivity. In the development of industrial robotic arms, camera sensors have many advantages due to their cost-effectiveness and small sizes. However, estimating object positions is a challenging problem, and it critically affects to the robustness of object manipulation functions. This paper proposes a method for estimating the 3D positions of objects, and it is applied to a pick-and-place task. A deep learning model is utilized to detect 2D bounding boxes in the image plane, and the pinhole camera model is employed to compute the object positions. To improve the robustness of measuring the 3D positions of objects, we analyze the effect of lens distortion and introduce a false positive filtering process. Experiments were conducted on a real-world scenario for moving medicine bottles by using a camera-based manipulator. Experimental results demonstrated that the distortion removal and false positive filtering are effective to improve the position estimation precision and the manipulation success rate.