• Journal of Korea Multimedia Society
• /
• v.21 no.12
• /
• pp.1425-1430
• /
• 2018

In this paper, we propose a CNN based deep learning algorithm for semantic segmentation of images. In order to improve the accuracy of semantic segmentation, we combined pixel level object classification and image level object classification. The image level object classification is used to accurately detect the characteristics of an image, and the pixel level object classification is used to indicate which object area is included in each pixel. The proposed network structure consists of three parts in total. A part for extracting the features of the image, a part for outputting the final result in the resolution size of the original image, and a part for performing the image level object classification. Loss functions exist for image level and pixel level classification, respectively. Image-level object classification uses KL-Divergence and pixel level object classification uses cross-entropy. In addition, it combines the layer of the resolution of the network extracting the features and the network of the resolution to secure the position information of the lost feature and the information of the boundary of the object due to the pooling operation.

• Journal of KIISE:Software and Applications
• /
• v.33 no.4
• /
• pp.427-439
• /
• 2006

The SBF (Simple Boundary Follower) is a boundary-following algorithm, and is used mainly for image recognition and presentation. The SBF is very popular because of its simplicity and efficiency in tracing the boundary of an object from an acquired binary image; however, it does have two drawbacks. First, the SBF cannot consistently process inner or inner-outer corners according to the follower's position and direction. Second, the SBF requires movement operations for the non-boundary pixels that are connected to boundary pixels. The MSBF (Modified Simple Boundary Follower) has a diagonal detour step for preventing inner-outer corner inconsistency, but is still inconsistent with inner-corners and still requires extra movement operations on non-boundary pixels. In this paper, we propose the ISBF (Improved Simple Boundary Follower), which solves the inconsistencies and reduces the extra operations. In addition, we have classified the tour maps by paths from a current boundary pixel to the next boundary pixel and have analyzed SBF, MSBF, and ISBF. We have determined that the ISBF has no inconsistency issues and reduces the overall number of operations.

• Journal of Korea Game Society
• /
• v.10 no.4
• /
• pp.63-71
• /
• 2010

In this paper, we propose a boundary based marker binary coding method for augmented reality games, which enables the marker-area to be binary coded well in any lighting environments. First, it detects the boundary after transforming an original marker image to a gray scale image, and it executes 4 way pixel extensions for all boundary pixels in order to make the boundary to closed area. Next, for all boundary pixels it compares the brightness of right and left ones of each pixel and allocates black for the lower side and white for the higher side by filling inside area thru the seeded region growing. Experimental results showed that our proposed method produces a good binary marker image recognizable in various light environments. In addition, it showed the possibility of real-time calculation by considering the result of operation speed which is 51 fps for VGA image.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.4
• /
• pp.381-387
• /
• 2002

In this paper, an efficient method to find shot boundaries in the MPEG video stream data is proposed. For this purpose, we first assume that the histogram difference value(HDV) and pixel difference value(PDV) as an one dimensional signal and apply the median filter to these signals. The output of the median filter is subtracted from the original signal to produce the median filtered difference(MFD). The MFD is a criterion of shot boundary. In addition a neural network is employed and trained to find exactly cut boundary. The proposed algorithm shows that the cut boundaries are well extracted, especially in a dynamic video.

• The Journal of the Convergence on Culture Technology
• /
• v.4 no.4
• /
• pp.361-366
• /
• 2018

For each character constituting learning data, different fonts are added in pixel unit to create MASK, and then pixel values belonging to the MASK are divided into three groups. The experimental data are modified into skeletal forms, and then regional boundary operation is used to create a boundary that distinguishes the background region adjacent to the skeleton of the character from the background of the modified experimental data. Discordance values between the modified experimental data and the MASKs are calculated, and then the MASK with the minimum value is found. This MASK is selected as a finally recognized result for the given experiment data. The recognition algorithm using skeleton of the character and the regional boundary operation can easily extend the learning data set by adding new fonts to the given learning data, and also it is simple to implement, and high character recognition rate can be obtained.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.34S no.5
• /
• pp.35-44
• /
• 1997

In this paper, a an algorithm that detects the endocardial boundary, expanding the region from endocardial cavity using fuzzy inference, is proposed. This algorithm decides the ventricular cavity by fuzzy inference in process of searching each pixel from the inside of left ventricle in echocardial image and expands it. Uncertainty and fuzziness exists in decision of endocardial boundary. Therefore, we convert the lingustic representation of mean, standard deviation, and threshold value that are characteristic variables of endocardial boundary to fuzzy input and output variables. And, we extract proposed method is robuster to noise than radial searching method that is highly dependent on center position. To prove the similarity of detected boundary by fuzzy nference, we used the measures of SIZE, correlation coefficient, MSD, and RMSE and had acquired reasonable results.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.8
• /
• pp.16-22
• /
• 1999

In this paper, an efficient algorithm for estimating real edge locations to subpixel values is described. Digital images are acquired by projection into image plane and sampling process. However, most of real edge locations are lost in this process, which causes low measurement accuracy. For accurate measurement, we propose an algorithm which estimates the real boundary between two adjacent pixels in digital imagery, with subpixel accuracy. We first define 1D edge operator based on the moment invariant. To extend it to 2D data, the edge orientation of each pixel is estimated by the LSE(Least Squares Error)line/circle fitting of a set of pixels around edge boundary. Then, using the pixels along the line perpendicular to the estimated edge orientation the real boundary is calculated with subpixel accuracy. Experimental results using real images show that the proposed method is robust in local noise, while maintaining low measurement error.

• Proceedings of the IEEK Conference
• /
• 2007.07a
• /
• pp.327-328
• /
• 2007

A contour shape matching based pixel motion estimation is proposed. The pixel motion information is very useful to compensate the motion artifact generated at the specific gray level contours in the moving image for subfield gray-scale display devices. In this motion estimation method, the gray level boundary contours are extracted from the input image. Then using contour shape matching, the most similar contour in next frame is found, and the contour is divided into segment unit. The pixel motion vector is estimated from the displacement of the each segment in the contour by segment matching. From this method, more precise motion vector can be estimated and this method is more robust to image motion with rotation or from illumination variations.

• The Journal of the Korea Contents Association
• /
• v.16 no.7
• /
• pp.431-438
• /
• 2016

In this paper, we present an upsampling technique for depth map image using color and depth weights. First, we construct a high-resolution image using the bilinear interpolation technique. Next, we detect a common edge region using RGB color space, HSV color space, and depth image. If an interpolated pixel belongs to the common edge region, we calculate weighting values of color and depth in $3{\times}3$ neighboring pixels and compute the cost value to determine the boundary pixel value. Finally, the pixel value having minimum cost is determined as the pixel value of the high-resolution depth image. Simulation results show that the proposed algorithm achieves good performance in terns of PSNR comparison and subjective visual quality.

• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.270-273
• /
• 2007

In this paper, a SAR-despeck1ing approach of adaptive iteration based a Bayesian model using the lognormal distribution for image intensity and a Gibbs random field (GRF) for image texture is proposed for noise removal of the images that are corrupted by multiplicative speckle noise. When the image intensity is logarithmically transformed, the speckle noise is approximately Gaussian additive noise, and it tends to a normal probability much faster than the intensity distribution. The MRF is incorporated into digital image analysis by viewing pixel types as states of molecules in a lattice-like physical system. The iterative approach based on MRF is very effective for the inner areas of regions in the observed scene, but may result in yielding false reconstruction around the boundaries due to using wrong information of adjacent regions with different characteristics. The proposed method suggests an adaptive approach using variable parameters depending on the location of reconstructed area, that is, how near to the boundary. The proximity of boundary is estimated by the statistics based on edge value, standard deviation, entropy, and the 4th moment of intensity distribution.