• Structural Engineering and Mechanics
• /
• v.69 no.4
• /
• pp.467-477
• /
• 2019

Digital image correlation (DIC) is now a popular and extensively used full-field metrology technique. In general, DIC is performed by using a turnkey solution offered by various manufacturers of DIC. In this paper, a simple and economical set-up for DIC is proposed which uses easily accessible digital single-lens reflex (DSLR) camera rather than industrial couple-charged device (CCD) cameras. The paper gives a description of aspects of carrying a DIC experiment which includes experimental set-up, specimen preparation, image acquisition and analysis. The details provided here will be helpful to carry DIC experiments without specialized DIC testing rig. To validate the responses obtained from proposed DIC set-up, tension and fatigue tests on specimens made of IS 2062 Gr. E300 steel are determined. Tensile parameters for a flat specimen and stress intensity factor for an eccentrically-loaded single edge notch tension specimen are evaluated from results of DIC experiment. Results obtained from proposed DIC experiments are compared with those obtained from conventional methods and are found to be in close agreement. It is also noted that the high resolution of DSLR allows the use of proposed approach for fracture characterization which could not be carried out with a typical turnkey DIC solution employing a camera of 2MP resolution.

• Journal of the Korean Society of Visualization
• /
• v.19 no.1
• /
• pp.28-35
• /
• 2021

In this study, an image processing method for the measurement of two-phase bubbly flow is developed. Shadowgraphy images obtained by high-speed camera are used for analysis. Some bubbles are generated as single unit and others are overlapped or clustered. Single bubbles can be easily analyzed using parameters such as bubble shape, centroid, and area. But overlapped bubbles are difficult to transform clustered bubbles into segmented bubbles. Several approaches were proposed for the bubble segmentation such as Hough transform, connection point method and watershed. These methods are not enough for bubble segmentation. In order to obtain the size distribution of bubbles, we present a method of splitting overlapping bubbles using watershed and approximating them to ellipse. There is only 5% error difference between manual and automatic analysis. Furthermore, the error can be reduced down to 1.2% when a correction factor is used. The ellipse fitting algorithm developed in this study can be used to measure bubble parameters accurately by reflecting the shape of the bubbles.

• Nuclear Engineering and Technology
• /
• v.54 no.10
• /
• pp.3943-3948
• /
• 2022

Single-photon emission computed tomography is one of the reliable pin-by-pin verification techniques for spent-fuel assemblies. One of the challenges with this technique is to increase the total fuel assembly verification speed while maintaining high verification accuracy. The aim of the present study, therefore, was to develop an artificial intelligence (AI) algorithm-based tomographic image analysis technique for partial-defect verification of fuel assemblies. With the Monte Carlo (MC) simulation technique, a tomographic image dataset consisting of 511 fuel-rod patterns of a 3 × 3 fuel assembly was generated, and with these images, the VGG16, GoogLeNet, and ResNet models were trained. According to an evaluation of these models for different training dataset sizes, the ResNet model showed 100% pattern estimation accuracy. And, based on the different tomographic image qualities, all of the models showed almost 100% pattern estimation accuracy, even for low-quality images with unrecognizable fuel patterns. This study verified that an AI model can be effectively employed for accurate and fast partial-defect verification of fuel assemblies.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.5
• /
• pp.259-265
• /
• 2024

In this paper, we propose an automatic display defect detection system using image processing. The existing inspection by operators has the disadvantage that human errors may occur due to the operator's skill level, fatigue, etc., and that standardization and quantification are difficult. It also has disadvantages such as the limited inspection speed and the cost of the operator's education. The proposed system automatically detects various display defects through image processing algorithms. It was developed based on the Jetson Nano, one of the most popular SBCs (Single Board Computers), and it has a conveyor belt to automatically moves the display to the inspection position. By providing a human machine interface (HMI), the operator can check the inspection information in real time, and control the inspection flow. By storing the inspection results as a log file, the operator can check the inspection results at any time, such as the time taken to perform each algorithm and the location of the detected defects. In addition, a multi-threaded structure was adopted. The camera's operations and inspection algorithms are executed in parallel in different threads, which can shorten the inspection time compared to the system based on a single-threaded structure. The experimental results prove the defect detection capability of the system and the efficiency of the inspection time.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.368-391
• /
• 2018

Face recognition (FR) with a single sample per person (SSPP) is common in real-world face recognition applications. In this scenario, it is hard to predict intra-class variations of query samples by gallery samples due to the lack of sufficient training samples. Inspired by the fact that similar faces have similar intra-class variations, we propose a virtual sample generating algorithm called k nearest neighbors based virtual sample generating (kNNVSG) to enrich intra-class variation information for training samples. Furthermore, in order to use the intra-class variation information of the virtual samples generated by kNNVSG algorithm, we propose image set based multimanifold discriminant learning (ISMMDL) algorithm. For ISMMDL algorithm, it learns a projection matrix for each manifold modeled by the local patches of the images of each class, which aims to minimize the margins of intra-manifold and maximize the margins of inter-manifold simultaneously in low-dimensional feature space. Finally, by comprehensively using kNNVSG and ISMMDL algorithms, we propose k nearest neighbor virtual image set based multimanifold discriminant learning (kNNMMDL) approach for single sample face recognition (SSFR) tasks. Experimental results on AR, Multi-PIE and LFW face datasets demonstrate that our approach has promising abilities for SSFR with expression, illumination and disguise variations.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.28 no.3
• /
• pp.294-300
• /
• 1995

In aquaculture industrial field, an automatic management and control system is needed to cope with the difficulties such as expensive wage, ripe age of management worker and risk according to the unexpected change of environmental conditions in the aquarium. This paper introduces an automatic aquarium monitoring and control system. The system is developed using PC single board computer. A PC can be connected to multi-single hoard computers, and the communication between PC and single board computers is based on RS-422/485 interfacing method. The physical data of pH, DO, temperature and water level etc. are real-timely treated in the single board computer though individual transducers, transfered to the main monitoring PC through RS-422/485 communication, and those data are graphically shown on the PC monitor. Furthermore, the environmental circumstance can be monitored through the image processing system, and the emergency system can be operated under the condition of environmental incident such as electric power stoppage, DO deficiency, pump shut down and low level water etc.

• Journal of Sensor Science and Technology
• /
• v.24 no.6
• /
• pp.359-363
• /
• 2015

This paper proposes an algorithm that reduces the conversion time of a single-slope A/D converter (SSADC) that has n-bit resolution, which typically is limited by conversion time taking up to $2^n$ clock cycles for an operation. To improve this situation, we have researched a novel hybrid-type A/D converter that consists of a pseudo-pipeline A/D converter and a conventional SSADC. The pseudo-pipeline A/D converter, using a single-stage of analog components, determines the most significant bits (MSBs) or upper bits and the conventional SSADC determines the remaining bits. Therefore, the modified SSADC, similar to the hybrid-type A/D converter, is able to significantly reduce the conversion time because the pseudo-pipeline A/D converter, which determines the MSBs (or upper bits), does not rely on a clock. The proposed A/D converter was designed using a $0.35-{\mu}m$ 2-poly 4-metal standard complementary metal oxide semiconductor (CMOS) technology process; additionally, its characteristics were simulated.

• Smart Structures and Systems
• /
• v.28 no.6
• /
• pp.811-825
• /
• 2021

High-voltage isolating switches play a paramount role in ensuring the safety of power supply systems. However, their exposure to outdoor environmental conditions may cause serious physical defects, which may result in great risk to power supply systems and society. Image processing-based methods have been used for anomaly detection. However, their accuracy is affected by numerous uncertainties due to manually extracted features, which makes the anomaly detection of isolating switches still challenging. In this paper, a vision-based anomaly detection method for isolating switches, which uses the rotational angle of the switch system for more accurate and direct anomaly detection with the help of deep learning (DL) and image processing methods (Single Shot Multibox Detector (SSD), improved frame differencing method, and Hough transform), is proposed. The SSD is a deep learning method for object classification and localization. In addition, an improved frame differencing method is introduced for better feature extraction and a hough transform method is adopted for rotational angle calculation. A number of experiments are conducted for anomaly detection of single and multiple switches using video frames. The results of the experiments demonstrate that the SSD outperforms the You-Only-Look-Once network. The effectiveness and robustness of the proposed method have been proven under various conditions, such as different illumination and camera locations using 96 videos from the experiments.

• Proceedings of the IEEK Conference
• /
• 2000.11d
• /
• pp.89-92
• /
• 2000

In this paper, an effective image restoration using Genetic Algorithm(GA) in wavelet transform region is proposed. First, a wavelet transform is used for decomposition of a blurred image with white Gaussian noise as a preprocessing of the proposed method. The wavelet transform decomposes a degraded image into a wavelet subband coefficient planes. In this wavelet transformed subband coefficient planes, three highest subbands is composed entirely of noise elements on a degraded image. So, these subbands are removed. And remained subbands except for the lowest subband are individually applied to GA. For the performance evaluation, the proposed method is compared with a conventional single GA algorithm and a conventional hybrid method of wavelet transform and GA for a Lenna image and a boat image. As an experimental result, the proposed algorithm is prior to a conventional methods as each PSNR 3.4dB, 1.3dB.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.4
• /
• pp.1714-1729
• /
• 2018

Change detection of remote sensing images is a profound challenge in the field of remote sensing image analysis. This paper proposes a novel change detection method for bitemporal remote sensing images based on feature fusion and fuzzy c-means (FCM). Different from the state-of-the-art methods that mainly utilize a single image feature for difference image construction, the proposed method investigates the fusion of multiple image features for the task. The subsequent problem is regarded as the difference image classification problem, where a modified fuzzy c-means approach is proposed to analyze the difference image. The proposed method has been validated on real bitemporal remote sensing data sets. Experimental results confirmed the effectiveness of the proposed method.