• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.3
• /
• pp.1086-1103
• /
• 2020

The automatic identification and classification of image-based weld defects is a difficult task due to the complex texture of the X-ray images of the weld defect. Several depth learning methods for automatically identifying welds were proposed and tested. In this work, four different depth convolutional neural networks were evaluated and compared on the 1631 image set. The concavity, undercut, bar defects, circular defects, unfused defects and incomplete penetration in the weld image 6 different types of defects are classified. Another contribution of this paper is to train a CNN model "RayNet" for the dataset from scratch. In the experiment part, the parameters of convolution operation are compared and analyzed, in which the experimental part performs a comparative analysis of various parameters in the convolution operation, compares the size of the input image, gives the classification results for each defect, and finally shows the partial feature map during feature extraction with the classification accuracy reaching 96.5%, which is 6.6% higher than the classification accuracy of other existing fine-tuned models, and even improves the classification accuracy compared with the traditional image processing methods, and also proves that the model trained from scratch also has a good performance on small-scale data sets. Our proposed method can assist the evaluators in classifying pipeline welding defects.

• Journal of Korea Multimedia Society
• /
• v.16 no.1
• /
• pp.108-118
• /
• 2013

Recently the 3D modeling techniques were developed rapidly due to rapid development of computer vision, computer graphics with the excellent performance of hardware. With the advent of a variety of 3D contents, 3D modeling technology becomes more in demand and it's quality is increased. 3D face models can be applied widely to such contents with high usability. In this paper, a 3D face modeling is attempted from a given single 2D frontal face image. To achieve the goal, we thereafter the feature points using AAM are extracted from the input frontal face image. With the extracted feature points we deform the 3D general model by 2-pass mesh warping, and also the depth extraction based on intensity values is attempted to. Throughout those processes, a universal 3D face modeling method with less expense and less restrictions to application environment was implemented and it's validity was shown through experiments.

• 한국지구물리탐사학회:학술대회논문집
• /
• 2008.10a
• /
• pp.93-96
• /
• 2008

We analyze a precise seabed feature around the Hupo Bank by using Multi-beam echosounder. Multi-beam echosounder system can observe the topography undulation according to the navigation of the survey ship by shooting wide beam. It is possible to embody a precision seabed feature because it can be make high density of incompletion depth sounding between survey lines. Through this survey, there is the Hupo Bank which is 84 km long, 1-15 km wide, 5.3-160 m deep in the center, at the west is moat, at the east is scarp and submarine canyon. The top of the Hupo Bank is the Wangdol reef that has 5.3 m in depth of water at least. Moat in survey area is 30 m long, and 30-40 m wide and has a depressed channel. The gap of depth of water in scarp is approximately 60 m and shows a characteristic of cuttig plane. Submarine canyon is 3.5 - 13.5 km wide.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.6
• /
• pp.819-824
• /
• 2021

As technologies related to image processing such as autonomous driving and CCTV develop, fog removal algorithms using a single image are being studied to improve the problem of image distortion. As a method of predicting fog density, there is a method of estimating the depth of an image by generating a depth map, and various fog features may be used as training data of the depth map. In addition, it is essential to implement a hardware capable of processing high-definition images in real time in order to apply the fog removal algorithm to actual technologies. In this paper, we implement NLCV (Normalize Local Coefficient of Variation), a feature of fog based on coefficient of variation, in hardware. The proposed hardware is an FPGA implementation of Xilinx's xczu7ev-2ffvc1156 as a target device. As a result of synthesis through the Vivado program, it has a maximum operating frequency of 479.616MHz and shows that real-time processing is possible in 4K UHD environment.

• The Journal of the Korea Contents Association
• /
• v.19 no.1
• /
• pp.125-131
• /
• 2019

A vanishing point is a point on an image to which parallel lines projected from a real space gather. A vanishing point in a road space provides important spatial information. It is possible to improve the position of an extracted lane or generate a depth map image using a vanishing point in the road space. In this paper, we propose a method of detecting vanishing points on images taken from a vehicle's point of view using Deep Neural Network (DNN) and Histogram of Oriented Gradient (HoG). The proposed algorithm is divided into a HoG feature extraction step, in which the edge direction is extracted by dividing an image into blocks, a DNN learning step, and a test step. In the learning stage, learning is performed using 2,300 road images taken from a vehicle's point of views. In the test phase, the efficiency of the proposed algorithm using the Normalized Euclidean Distance (NormDist) method is measured.

• Journal of KIISE:Software and Applications
• /
• v.35 no.10
• /
• pp.616-626
• /
• 2008

The face shape extracted by the depth values has different appearance as the most important facial feature information and the face images decomposed into frequency subband are signified personal features in detail. In this paper, we develop a method for recognizing the range face images by combining the multiple frequency domains for each depth image and depth fusion using fuzzy integral. For the proposed approach, the first step tries to find the nose tip that has a protrusion shape on the face from the extracted face area. It is used as the reference point to normalize for orientated facial pose and extract multiple areas by the depth threshold values. In the second step, we adopt as features for the authentication problem the wavelet coefficient extracted from some wavelet subband to use feature information. The third step of approach concerns the application of eigenface and Linear Discriminant Analysis (LDA) method to reduce the dimension and classify. In the last step, the aggregation of the individual classifiers using the fuzzy integral is explained for extracted coefficient at each resolution level. In the experimental results, using the depth threshold value 60 (DT60) show the highest recognition rate among the regions, and the depth fusion method achieves 98.6% recognition rate, incase of fuzzy integral.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.1
• /
• pp.132-148
• /
• 2013

Via the Internet, the information infrastructure of modern health care has already established medical information systems to share electronic health records among patients and health care providers. Data hiding plays an important role to protect medical images. Because modern medical devices have improved, high resolutions of medical images are provided to detect early diseases. The high quality medical images are used to recognize complicated anatomical structures such as soft tissues, muscles, and internal organs to support diagnosis of diseases. For instance, 16-bit depth medical images will provide 65,536 discrete levels to show more details of anatomical structures. In general, the feature of low utilization rate of intensity in 16-bit depth will be utilized to handle overflow/underflow problem. Nowadays, most of data hiding algorithms are still experimenting on 8-bit depth medical images. We proposed a novel reversible data hiding scheme testing on 16-bit depth CT and MRI medical image. And the peak point and zero point of a histogram are applied to embed secret message k bits without salt-and-pepper.

• Journal of the Korean Geotechnical Society
• /
• v.21 no.1
• /
• pp.115-121
• /
• 2005

At present, continual road constructions to connect from city to city are needed due to the geographical feature of Korea that about $70\%$ of the territory is mountainous area. Thus, the generation of large cut-slope has been inevitably formed. As a means of reinforcement on the cut-slope, in case of destructive disasters such as a snowstorm, pile embedment method is widely adopted. The pile embedment method is to resist possible move of soil by embedding piles from the surface to the immovable ground and then delivering the load from the piles to the immovable ground. In this study this writer analyzes the limitation of empirically used pile construction depth and its spacing through the numerical analysis. As a result, he suggests the most effective pile construction depth and space.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.9
• /
• pp.790-797
• /
• 2001

A simple computer vision technology to measure the middle-ranged depth with a mono camera and a plain mirror is proposed. The proposed system is structured with the rotating mirror in front of the fixed mono camera. In contrast to the previous stereo vision system in which the disparity of the closer object is larger than that of the distant object, the pixel movement caused by the rotating mirror is bigger for the pixels of the distant object in the proposed system. Being inspired by such distinguished feature in the proposed system, the principle of the depth measurement based on the relation of the pixel movement and the distance of object is investigated. Also, the factors to influence the precision of the measurement are analysed. The benefits of the proposed system are low price and less chance of occlusion. The robustness for practical usage is an additional benefit of the proposed vision system.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.12
• /
• pp.2943-2948
• /
• 2015

Run 3D reconstruction using a single camera, based on the information, we are advancing research on driving assistance apparatus or can be informed how to pass the obstacle existing ahead the driver. As a result depth information falls but it is possible to provide information that can pass through an obstacle on the straight. For 3D reconstruction by measuring the internal parameters, it calculates the Fundamental matrix and matching to find the feature points obtained by executing the triangulation on the basis of this. When the through experiments try to confirm the results, the depth information is present error information in the X and Y axes which can determine whether or not to pass through an obstacle has reliability.