• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.2
• /
• pp.75-82
• /
• 2000

This paper presents a method of detecting surface flaw of cold-rolled steel plate using image processing technique and a neural network classifier. The amount of steel plate surface image data is reduced by the wavelet transform. Features are extracted from the co-occurence matrix of the partial image corresponding to the low-frequency region, and a MLP neural network classifies into predetermined surface flaw categories. Simulations show the neural network classifier outperforms conventional vector quantization method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.12
• /
• pp.2271-2282
• /
• 2006

This paper is concerned with simulation issues arising in the PDE-based image restoration such as the total variation minimization(TVM) and its generalizations. In particular, we study the issues of staircasing and excessive dissipation of TVM-like smoothing operators. A strategy of scaling the algebraic system and a non-convex minimization are considered respectively for anti-staircasing and anti-diffusion. Furthermore, we introduce a variable constraint parameter to better preserve image edges. The resulting algorithm has been numerically verified to be efficient and reliable in denoising. Various numerical results are shown to confirm the claim.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.529-534
• /
• 2009

This paper reports a non-photorealistic rendering method for creating stream pattern from an input image. Our method extracts potential stream pattern in the given image. The proposed approach uses a shock filter based on a partial difference equation(PDE) which is implemented by applying a selective dilation and erosion processes. However, unlike the traditional first order solution to the PDE, we employ a second order scheme and compensate for the undesired diffusive effects caused by a viscosity form. The selection of dilation or erosion for a pixel is based on an edge detector computed from a structure tensor. By adding noises on to the input image, our method also can generate stream pattern even if there is less texture in some area. The experimental results show that the stream pattern is extracted very well.

• Proceedings of the Korean Society of Marine Engineers Conference
• /
• 2005.06a
• /
• pp.534-539
• /
• 2005

Particle image analyzer for measuring droplet size has been developed. Image processing technique was used with relaxation method. The morphological method based on partial curvature information of pre-processed images was adopted for recognition and separation of overlapped particles. The measurement results show that the present method may be reliable for the analysis of the size and distribution of droplets produced by water mist spay flow.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.11
• /
• pp.153-162
• /
• 1995

When vector quantization is used in low rate image coding (e.g., R<0.5), the primary problem is the tremendous computational complexity which is required to search the whole codebook to find the closest codevector to an input vector. Since the number of code vectors in a vector quantizer is given by an exponential function of the dimension. i.e., L=2$^{nR}$ where Rn. To alleviate this problem, a multiple shell structure of hypercube feature maps (MSSHFM) is proposed. A binary HFM of k-dimension is composed of nodes at hypercube vertices and a multiple shell architecture is constructed by surrounding the k-dimensional hfm with a (k+1)-dimensional HFM. Such a multiple shell construction of nodes inherently has a complete tree structure in it and an efficient partial search scheme can be applied with drastically reduced computational complexity, computer simulations of still image coding were conducted and the validity of the proposed method has been verified.

• Journal of Information Processing Systems
• /
• v.16 no.4
• /
• pp.991-1000
• /
• 2020

We propose a content-based image retrieval system that uses a combination of completed local binary pattern (CLBP) and color autocorrelogram. CLBP features are extracted on a multi-resolution multi-direction filtered domain of value component. Color autocorrelogram features are extracted in two dimensions of hue and saturation components. Experiment results revealed that the proposed method yields a lot of improvement when compared with the methods that use partial features employed in the proposed method. It is also superior to the conventional CLBP, the color autocorrelogram using R, G, and B components, and the multichannel decoded local binary pattern which is one of the latest methods.

• Smart Media Journal
• /
• v.5 no.2
• /
• pp.15-23
• /
• 2016

In this paper we propose the depth-poselets using body-part-poses and also propose the method to recognize the gesture. Since the gestures are composed of sequential poses, in order to recognize a gesture, it should emphasize to obtain the time series pose. Because of distortion and high degree of freedom, it is difficult to recognize pose correctly. So, in this paper we used partial pose for obtaining a feature of the pose correctly without full-body-pose. In this paper, we define the 16 gestures, a depth image using a learning image was generated based on the defined gestures. The depth poselets that were proposed in this paper consists of principal three-dimensional coordinates of the depth image and its depth image of the body part. In the training process after receiving the input defined gesture by using a depth camera in order to train the gesture, the depth poselets were generated by obtaining 3D joint coordinates. And part-gesture HMM were constructed using the depth poselets. In the testing process after receiving the input test image by using a depth camera in order to test, it extracts foreground and extracts the body part of the input image by comparing depth poselets. And we check part gestures for recognizing gesture by using result of applying HMM. We can recognize the gestures efficiently by using HMM, and the recognition rates could be confirmed about 89%.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.1
• /
• pp.1-7
• /
• 2023

An artificial intelligence (AI) method based on image deep learning is proposed to predict the entire displacement shape of a structure using the feature of partial displacements. The performance of the method was investigated through a structural test of a steel frame. An image-to-image regression (I2IR) training method was developed based on the U-Net layer for image recognition. In the I2IR method, the U-Net is modified to generate images of entire displacement shapes when images of partial displacement shapes of structures are input to the AI network. Furthermore, the training of displacements combined with the location feature was developed so that nodal displacement values with corresponding nodal coordinates could be used in AI training. The proposed training methods can consider correlations between nodal displacements in 3D space, and the accuracy of displacement predictions is improved compared with artificial neural network training methods. Displacements of the steel frame were predicted during the structural tests using the proposed methods and compared with 3D scanning data of displacement shapes. The results show that the proposed AI prediction properly follows the measured displacements using 3D scanning.

• Journal of radiological science and technology
• /
• v.45 no.3
• /
• pp.241-248
• /
• 2022

In this study, a recovery coefficient (RC) calculation was conducted that can correct the underestimation of the standardized uptake value (SUV) due to the partial volume effect (PVE) through phantom measurements and formulas. The experiment was conducted using a dynamic phantom capable of implement cranio-caudal movement at a respiratory rate of 15 times per minute along with the measured phantom experiment of the stopped state, and the RC of the moving state is calculated and compared. Ingenuity TF (Philips Healthcare, Netherland) was used as a positron emission tomography/computed tomography (PET/CT) device. PET-CT Phantom (Biodex Medical System, USA) was used as a phantom for measurement. A phantom image in a stationary state was acquired, and a moving phantom image was acquired using the AZ-733V Respiratory Phantom (Anzai Medical Co, Japan) capable of breathing movement in the cranio-caudal direction under the same acquisition parameters. For RC calculation, the sphere maximum radioactivity concentration and the background mean radioactivity concentration of the acquired images were measured, and the initially determined sphere and background radioactivity concentrations were calculated. The calculated RC was 0.08 to 0.72. The size of sphere smaller, it was confirmed that the RC reduced. And the RC in the moving state reduced than in the stationary state. As a result of this study, the change of the RC was confirmed according to the size of spheres and the phantom moving. Using the RC derived by implement movement of breathing with the respiratory phantom, it is possible to considering correction of underestimated SUV by the partial volume effect of PET images and the patient movements.

• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.36D no.2
• /
• pp.62-70
• /
• 1999

This paper introduces a possibility of a low-cost, high-resolution fingerprint sensor chip. The test chip is composed of $64{\times}256$ sensing cells(chip size : $2.7mm{\times}10.8mm$). A new detection circuit of charge sharing is proposed, which eliminates the influences of internal parasitic copacitances. This the reduced sensing-capacitor size enables a high resolution of 600dpi, using even conventional 0.6${\mu}m$ CMOS process. The partial fingerprint image captured therefrom are synthesized into a full fingerprint image with a image synthesis algorithm. The problems and possibilities of image synthesis technique are also analyzed and discussed.