• Journal of Mechanical Science and Technology
• /
• v.17 no.3
• /
• pp.409-421
• /
• 2003

The hierarchical recursive local-correlation PIV algorithm with CBC (correlation based correction) method was employed to increase the spatial resolution of PIV results and to reduce error vectors. The performance of this new PIV algorithm was tested using synthetic images, PIV standard images of Visualization Society of Japan, real flows including ventilation flow inside a vehicle passenger compartment and wake behind a circular cylinder with riblet surface. As a result, most spurious vectors were suppressed by employing the CBC method, the hierarchical recursive correlation algorithm improved the sub-pixel accuracy of PIV results by decreasing the interrogation window size and Increased spatial resolution significantly. However, with recursively decreasing of interrogation window size, the SNR (signal-to-noise ratio) in the correlation plane was decreased and number of spurious vectors was increased. Therefore, compromised determination of optimal interrogation window size is required for given flow images, the performance of recursive algorithm is also discussed from a viewpoint of recovery ratio and error ratio in the paper.

• Agricultural and Biosystems Engineering
• /
• v.3 no.1
• /
• pp.18-28
• /
• 2002

Neural network based models were developed and evaluated for predicting corn yield from aerial images based on 1998 and 1994 image data. The model used images in multi-spectral bands such as R, G, B, and IR (Red, Green, Blue and Infrared). The inputs to the neural network consisted of mean and standard deviation of multispectral bands of the aerial images. Performances of several neural network architectures using back-propagation with momentum were compared. The maximum yield prediction accuracy obtained was 97.81%. The BPNN model prediction accuracy could be enhanced by using more number of observations to the model, other data transformation techniques, or by performing optical calibration of the aerial image.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.896-899
• /
• 1996

A new algorithm for planning a collision free path is developed based on linear parametric curve. A collision-free path is viewed as a connected space curve in which the path consists of two straight curve connecting start to target point. A single intermediate connection point is considered in this paper and is used to manipulate the shape of path by organizing the control point in polar coordinate (.theta.,.rho.). The algorithm checks interference with obstacles, defined as GM (Geometry Mapping), and maps obstacles in Euclidean Space into images in CPS (Connection Point Space). The GM for all obstacles produces overlapping images of obstacle in CPS. The clear area of CPS that is not occupied by obstacle images represents collision-free paths in Euclidean Space. Any points from the clear area of CPS is a candidate for a collision-free path. A simulation of GM for number of cases are carried out and results are presented including mapped images of GM and performances of algorithm.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.29B no.10
• /
• pp.78-89
• /
• 1992

In this paper, a multiresolution edge detction algorithm for speckle images is proposed. Due to the signal dependency of speckle images, the number of edge points detected depends on the local average intensity. Thus the edge detection method independent of the average intensity is required to detect properly real significant changes in an original signal. In the proposed method, candidate area is first selected based on the statistical propeties of speckle images,i.e., based on the busyness measure such as the CoV(coefficient of variation) and the difference between the real and theoretical CDF(cumulative density function). Then the real edges are extracted in a multiresolution environment. Computer simulation with test images shows that the proposed method reduces significantly false edges in relatively homogeneous areas while detects fine details properly.

• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.5
• /
• pp.558-567
• /
• 1999

X-ray laminography and DT(digital tomosynthesis) that can form a cross-sectional image of 3-D objects promis to be good solutions for inspecting interior defects of industrial products. DT is a kind of laminography technique and the difference is in the fact that it synthesizes the several projected images by use of the digitized memory and computation. The quality of images acquired from the DT system varies according to image synthesizing methods, the number of images used in image synthesizing, and X-ray projection angles. In this paper, a new image synthesizing method named 'log-root method' is proposed to get clear and accurate cross-sectional images, which can reduce both artifact and blurring generated by materials out of focal plane. To evaluate the quality of cross-sectional images, two evaluating criteria : (1) shape accuracy and (2) clearness of the cross-sectional images are defined. Based on these criteria, a series of simulations are performed, and the results show the superiority of the new synthesizing method over the existing ones such as averaging and minimum methods.

• Journal of Information Processing Systems
• /
• v.16 no.5
• /
• pp.1183-1195
• /
• 2020

Visual saliency detection is an essential task because it is an important part of various vision-based applications. There are many techniques for saliency detection in color images. However, the number of methods for saliency detection in infrared images is limited. In this paper, we introduce a simple approach for saliency detection in infrared images based on the thresholding technique. The input image is thresholded into several Boolean maps, and an initial saliency map is calculated as a weighted sum of the created Boolean maps. The initial map is further refined by using thresholding, morphology operation, and a Gaussian filter to produce the final, high-quality saliency map. The experiment showed that the proposed method has high performance when applied to real-life data.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.615-618
• /
• 2019

Visual saliency detection is an important part in various vision-based applications. There are a myriad of techniques for saliency detection in color images. However, the number of methods for saliency detection in infrared images is inadequate. In this paper, we introduce a simple approach for pedestrian identification in infrared images using saliency. The input image is thresholded into several Boolean maps, an initial saliency map is then calculated as a weighted sum of created Boolean maps. The initial map is further refined by using thresholding, morphology operation, and Gaussian filter to produce the final, high-quality saliency map. The experiment showed that the proposed method produced high performance results when applied to real-life data.

• Journal of Biomedical Engineering Research
• /
• v.45 no.3
• /
• pp.118-127
• /
• 2024

Commonly deep learning methods for enhancing the quality of medical images use unpaired dataset due to the impracticality of acquiring paired dataset through commercial imaging system. In this paper, we propose a supervised learning method to enhance the quality of ultrasound images. The U-net model is designed by incorporating a divide-and-conquer approach that divides and processes an image into four parts to overcome data shortage and shorten the learning time. The proposed model is trained using paired dataset consisting of 828 pairs of low-quality and high-quality images with a resolution of 512x512 pixels obtained by varying the number of channels for the same subject. Out of a total of 828 pairs of images, 684 pairs are used as the training dataset, while the remaining 144 pairs served as the test dataset. In the test results, the average Mean Squared Error (MSE) was reduced from 87.6884 in the low-quality images to 45.5108 in the restored images. Additionally, the average Peak Signal-to-Noise Ratio (PSNR) was improved from 28.7550 to 31.8063, and the average Structural Similarity Index (SSIM) was increased from 0.4755 to 0.8511, demonstrating significant enhancements in image quality.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.24 no.1
• /
• pp.12-21
• /
• 2021

For successful automatic target recognition(ATR) with synthetic aperture radar(SAR) imagery, SAR target images of the database should have the identical or highly similar resolution with those collected from SAR sensors. However, it is time-consuming or infeasible to construct the multiple databases with different resolutions depending on the operating SAR system. In this paper, an approach for resolution conversion of SAR target images is proposed based on conditional generative adversarial network(cGAN). First, a number of pairs consisting of SAR target images with two different resolutions are obtained via SAR simulation and then used to train the cGAN model. Finally, the model generates the SAR target image whose resolution is converted from the original one. The similarity analysis is performed to validate reliability of the generated images. The cGAN model is further applied to measured MSTAR SAR target images in order to estimate its potential for real application.

• Proceedings of the Korea Contents Association Conference
• /
• 2009.05a
• /
• pp.870-874
• /
• 2009

Genetic algorithms can be effectively used when generating abstract images in an automatic way. However, managing huge number of automatically generated images has been problematic without sufficient managing mechanisms. This paper presents effective classification scheme for the abstract Affine images using form, emotion and color facets, and implements image databases.