• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.05a
• /
• pp.215-218
• /
• 2013

In this paper, we proposed a method for enhancement of color images with sky regions. The input image is converted into HSV space and then sky and non-sky regions are separated. For sky region, saturation enhancement is performed for each pixel based on the enhancement factor calculated from the average saturation of its local neighborhood. On the other hand, for the non-sky region, the enhancement is applied only on the luminance value (V) component of the HSV color image, which is performed in two steps. The luminance enhancement, which is also called as dynamic range compression, is carried out using nonlinear transfer function. Again, each pixel is further enhanced for the adjustment of the image contrast depending upon the center pixel and its neighborhood pixel values. At last, the original H and V component image and enhanced S component image for the sky region, and original H and S component image and enhanced V component image for the non-sky region are converted back to RGB image.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.4
• /
• pp.1521-1537
• /
• 2021

Through information hiding technique, secret message can be hidden in pictures. Stego-image quality and hiding capacity are two important metrics for information hiding. To enhance these metrics, many schemes were proposed by scholars in recent years. Some of them are effective and successful, but there is still a room for further improvement. A high capacity information hiding scheme (PAMO, Pixel-value Adjustment with Modulus Operation Algorithm) is introduced in this paper. PAMO scheme uses pixel value adjustment with modulus operation to hide confidential data in cover-image. PAMO scheme and some referenced schemes are implemented in Python and experiments are carried out to evaluate their performance. In the experiments, PAMO scheme shows better performance than other methods do. When secret message length is less than 72000 bits, the highest hiding capacity of PAMO can reach 7 bits per pixel, at the same time the PSNR of stego-images is greater than 30 dB.

• Proceedings of the KSRS Conference
• /
• 2004.10a
• /
• pp.359-362
• /
• 2004

In multi-dimensional image, ICA-based feature extraction algorithm, which is proposed in this paper, is for the purpose of detecting target feature about pixel assumed as a linear mixed spectrum sphere, which is consisted of each different type of material object (target feature and background feature) in spectrum sphere of reflectance of each pixel. Landsat ETM+ satellite image is consisted of multi-dimensional data structure and, there is target feature, which is purposed to extract and various background image is mixed. In this paper, in order to eliminate background features (tidal flat, seawater and etc) around target feature (laver farm) effectively, pixel spectrum sphere of target feature is projected onto the orthogonal spectrum sphere of background feature. The rest amount of spectrum sphere of target feature in the pixel can be presumed to remove spectrum sphere of background feature. In order to make sure the excellence of feature extraction method based on ICA, which is proposed in this paper, laver farm feature extraction from Landsat ETM+ satellite image is applied. Also, In the side of feature extraction accuracy and the noise level, which is still remaining not to remove after feature extraction, we have conducted a comparing test with traditionally most popular method, maximum-likelihood. As a consequence, the proposed method from this paper can effectively eliminate background features around mixed spectrum sphere to extract target feature. So, we found that it had excellent detection efficiency.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.1
• /
• pp.120-127
• /
• 2010

This paper presents a SR method using adaptive interpolation based on local gradient features to obtain a high quality SR image. In this method, the distance between the interpolated pixel and the neighboring valid pixel is considered by using local gradient properties. The interpolation coefficients take the local gradient of the LR images into account. The smaller the local gradient of a pixel is, the more influence it should have on the interpolated pixel. And the median filter is finally applied to reduce the blurring and noise of the interpolated HR image. Experiment results show the effectiveness of the proposed method in comparison with other methods, especially in the edge areas of the images.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.3
• /
• pp.712-717
• /
• 2013

In this paper, we propose a performance-enhanced object recognition by using nonlinear 3D correlator based on pixel restoration. In the proposed method, elemental images of the 3D target that are partially occluded by a foreground object are picked up and transformed into sub-images. By using the block-matching algorithm, the occluded target regions of each sub-image are estimated and removed. After that, the missing pixels in each sub-image are reestablished by using the pixel-restoration method. Finally, through the nonlinear cross-correlations between the reconstructed reference and the target plane images, the improved object recognition can be performed. To show the feasibility of the proposed method, some preliminary experiments are carried out and results are presented by comparing the conventional method.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.43 no.6 s.348
• /
• pp.1-8
• /
• 2006

In this paper, we designed a pixel circuit for AMOLED display based on organic thin film transistors and analyzed the operation with SPICE simulation. First, we theoretically designed the pixel circuit with the result of layout for fabricating $32\times32$ AMOLED panel, TFT W/L and capacitance of storage capacitor. And we simulated the designed pixel circuit using HSPICE for analyzing electrical performance. As a result of simulation, we identified the possibility of AMOLED display based on OTFTs.

• Journal of KIISE:Software and Applications
• /
• v.29 no.9
• /
• pp.631-644
• /
• 2002

In this paper efficient algorithms for content-based video retrieval using motion information are proposed, including temporal scale-invariant retrieval and temporal scale-absolute retrieval. In temporal scale-invariant video retrieval, the distance transformation is performed on each trail image in database. Then, from a given que교 trail the pixel values along the query trail are added in each distance image to compute the average distance between the trails of query image and database image, since the intensity of each pixel in distance image represents the distance from that pixel to the nearest edge pixel. For temporal scale-absolute retrieval, a new coding scheme referred to as Motion Retrieval Code is proposed. This code is designed to represent object motions in the human visual sense so that the retrieval performance can be improved. The proposed coding scheme can also achieve a fast matching, since the similarity between two motion vectors can be computed by simple bit operations. The efficiencies of the proposed methods are shown by experimental results.

• Journal of Multimedia Information System
• /
• v.8 no.4
• /
• pp.203-210
• /
• 2021

Deep leaning convolutional neural networks (CNN) have successfully been applied to image super-resolution (SR). Despite their great performances, SR techniques tend to focus on a certain upscale factor when training a particular model. Algorithms for single model multi-scale networks can easily be constructed if images are upscaled prior to input, but sub-pixel convolution upsampling works differently for each scale factor. Recent SR methods employ multi-scale and multi-path learning as a solution. However, this causes unshared parameters and unbalanced parameter distribution across various scale factors. We present a multi-scale single-path upsample module as a solution by exploiting the advantages of sub-pixel convolution and interpolation algorithms. The proposed model employs sub-pixel convolution for the highest scale factor among the learning upscale factors, and then utilize 1-dimension interpolation, compressing the learned features on the channel axis to match the desired output image size. Experiments are performed for the single-path upsample module, and compared to the multi-path upsample module. Based on the experimental results, the proposed algorithm reduces the upsample module's parameters by 24% and presents slightly to better performance compared to the previous algorithm.

• Journal of the Korean Society of Visualization
• /
• v.21 no.1
• /
• pp.18-25
• /
• 2023

In these days, high dimensional data prediction technology based on neural network shows compelling results in many different kind of field including engineering. Especially, a lot of variants of convolution neural network are widely utilized to develop pixel level prediction model for high dimensional data such as picture, or physical field value from the sensors. In this study, velocity vector field of ideal flow on ship surface is estimated on pixel level by Unet. First, potential flow analysis was conducted for the set of hull form data which are generated by hull form transformation method. Thereafter, four different neural network with a U-shape structure were conFig.d to train velocity vectors at the node position of pre-processed hull form data. As a result, for the test hull forms, it was confirmed that the network with short skip-connection gives the most accurate prediction results of streamlines and velocity magnitude. And the results also have a good agreement with potential flow analysis results. However, in some cases which don't have nothing in common with training data in terms of speed or shape, the network has relatively high error at the region of large curvature.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.9
• /
• pp.2483-2504
• /
• 2023

Most existing low-light enhancement algorithms either use a large number of training parameters or lack generalization to real-world scenarios. This paper presents a novel lightweight and robust pixel-wise polynomial approximation-based deep network for low-light image enhancement. For mapping the low-light image to the enhanced image, pixel-wise higher-order polynomials are employed. A deep convolution network is used to estimate the coefficients of these higher-order polynomials. The proposed network uses multiple branches to estimate pixel values based on different receptive fields. With a smaller receptive field, the first branch enhanced local features, the second and third branches focused on medium-level features, and the last branch enhanced global features. The low-light image is downsampled by the factor of 2b-1 (b is the branch number) and fed as input to each branch. After combining the outputs of each branch, the final enhanced image is obtained. A comprehensive evaluation of our proposed network on six publicly available no-reference test datasets shows that it outperforms state-of-the-art methods on both quantitative and qualitative measures.