• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.195-206
• /
• 2022

Crack detection is essential for inspection of existing structures and crack segmentation based on deep learning is a significant solution. However, datasets are usually one of the key issues. When building a new dataset for deep learning, laborious and time-consuming annotation of a large number of crack images is an obstacle. The aim of this study is to develop an approach that can automatically select a small portion of the most informative crack images from a large pool in order to annotate them, not to label all crack images. An active learning method with difficulty learning mechanism for crack segmentation tasks is proposed. Experiments are carried out on a crack image dataset of a steel box girder, which contains 500 images of 320×320 size for training, 100 for validation, and 190 for testing. In active learning experiments, the 500 images for training are acted as unlabeled image. The acquisition function in our method is compared with traditional acquisition functions, i.e., Query-By-Committee (QBC), Entropy, and Core-set. Further, comparisons are made on four common segmentation networks: U-Net, DeepLabV3, Feature Pyramid Network (FPN), and PSPNet. The results show that when training occurs with 200 (40%) of the most informative crack images that are selected by our method, the four segmentation networks can achieve 92%-95% of the obtained performance when training takes place with 500 (100%) crack images. The acquisition function in our method shows more accurate measurements of informativeness for unlabeled crack images compared to the four traditional acquisition functions at most active learning stages. Our method can select the most informative images for annotation from many unlabeled crack images automatically and accurately. Additionally, the dataset built after selecting 40% of all crack images can support crack segmentation networks that perform more than 92% when all the images are used.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.4
• /
• pp.372-378
• /
• 2014

Noise in biomedical X-ray image degrades the quality so that it might causes to decrease the accuracy of diagnosis. Especially the noise reduction techniques is quite essential for low-dose biomedical X-ray images obtained from low radiation power in order to protect patients, because their noise level is usually high to well discriminate objects. This paper proposes an efficient method to remove the noise in low-dose X-ray images while preserving the edges with diverse resolutions. In the proposed method, a noisy image is at first decomposed into several images with different resolutions in pyramidal representation, then the stable map of edge confidence is obtained from each of analyzed image using a fuzzy logic-based edge detector. This map is used to adaptively determine the parameter for guided filters, which eliminate the noise while preserving edges in the corresponding image. The filtered images in the pyramid are extended and synthesized into a resulted image using interpolation technique. The superiority of proposed method compared to the median, bilateral, and guided filters has been experimentally shown in terms of noise removal and edge preserving properties.

• Journal of Broadcast Engineering
• /
• v.7 no.1
• /
• pp.45-54
• /
• 2002

This paper proposes a video compression algorithm using characteristics of wavelet coefficients. The proposed algorithm can provide lowed bit rate and faster running time while guaranteeing the reconstructed image qualify by the human virtual system. In this approach, each video sequence is decomposed into a pyramid structure of subimages with various resolution to use multiresolution capability of discrete wavelet transform. Then similarities between two neighboring frames are obtained from a low-frequency subband which Includes an important information of an image and motion informations are extracted from the similarity criteria. Four legion selection filters are designed according to the similarity criteria and compression processes are carried out by encoding the coefficients In preservation legions and replacement regions of high-frequency subbands. Region selection filters classify the high-frequency subbands Into preservation regions and replacement regions based on the similarity criteria and the coefficients In replacement regions are replaced by that of a reference frame or reduced to zero according to block-based similarities between a reference frame and successive frames. Encoding is carried out by quantizing and arithmetic encoding the wavelet coefficients in preservation regions and replacement regions separately. A reference frame is updated at the bottom point If the curve of similarity rates looks like concave pattern. Simulation results show that the proposed algorithm provides high compression ratio with proper Image quality. It also outperforms the previous Milton's algorithm in an Image quality, compression ratio and running time, leading to compression ratio less than 0.2bpp. PSNR of 32 dB and running tome of 10ms for a standard video image of size 352${\times}$240 pixels.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.10
• /
• pp.948-956
• /
• 2014

Today, Intelligent Vehicle Detection System seeks to reduce the negative factors, such as accidents over to get the traffic information of existing system. This paper proposes detection algorithm for the illegal U-turn vehicles which can cause critical accident among violations of road traffic laws. We predicted that if calculated optical flow vectors were shown on the illegal U-turn path, they would be cause of the illegal U-turn vehicles. To reduce the high computational complexity, we use the algorithm of pyramid Lucas-Kanade. This algorithm only track the key-points likely corners. Because of the high computational complexity, we detect center lane first through the color information and progressive probabilistic hough transform and apply to the around of center lane. And then we select vectors on illegal U-turn path and calculate reliability to check whether vectors is cause of the illegal U-turn vehicles or not. Finally, In order to evaluate the algorithm, we calculate process time of the type of algorithm and prove that proposed algorithm is efficiently.

• Journal of the Korean Society of Radiology
• /
• v.14 no.5
• /
• pp.635-642
• /
• 2020

The purpose of this study was to analysis the effects of shielding area of field configuration with changing of sensitivity and density on tube current (milliampere-seconds, mAs) and image quality in automatic exposure control (AEC) system. The equipment used a digital radiography device (Digital Diagnost, Philips, Netherlands), which has a integral type with an X-ray tube and an indirect digital detector. The AEC system conditions were consisted of 9 setting environments, that mode changing of the sensitivity (S200, S400, S800) and the density (+2.5, 0, -2.5). The tube current evaluated automatically exposed mAs under 81 combination conditions crossed by AEC conditions in fixed at 40 kVp. The image quality evaluated the radiographic images that selected valid images by visual assessment the radiographic images of the self-produced conical pyramid phantom and then measured their signal to noise ratio (SNR). As a result, the maximum tube current was 60.0 mAs that automatically exposed conditions were the 100% of shielding area and the sensitivity of S200 and the density of +2.5. The minimum tube current was 0.9 mAs with non-shielding area and the sensitivity of S800 and the density of -2.5. When the shielded area 0% with the sensitivity of S200 and the density of +2.5, the maximum SNR was the highest as 25.2. But when the shielded area 25% with the sensitivity of S800 and the density of -2.5, the minimum SNR was the lowest as 4.7.

• Journal of radiological science and technology
• /
• v.38 no.3
• /
• pp.261-266
• /
• 2015

Intravenous urography (IVU) for urolithiasis is a radiologic examination to diagnosis stone in the ureter path using iodine contrast media, which is radioopacity material. The method includes compression on the upper iliac crest. The compression band prevents outlet of the contrast media through the bladder and enables easier movement to upper urinary tract. This usage depends on the policy of a hospital. Therefore, this study aimed to review and compare the characteristic of progress of contrast media either in compression and non-compression. The retrospective image measurement on 60 cases of intravenous pyelography was conducted at a hospital with the identical type and amount of contrast media as well as criteria for testing. Image measurement was limited to 5 minutes clip, which is optimal for progress of contrast media depending on usage of the compression band. Also, anatomical regions were set as following: "RP" is from renal pyramid to renal pelvis, "PL" is from renal pelvis to lumbar three endplate, and "IU" and "IL" for upper and lower parts from both iliac crests. Analysis has been conducted through the statistical method based on Fisher's Exact Test to find if there are differences of distribution with the anatomical regions with compression or no compression. It has been confirmed that there is no statistical significant difference as the video measurement on 30 cases of compression and non-compression group respectively resulted in P value of 0.580 from left and 0.711 from right (both 0.960). Therefore, it has been concluded that application of a compression band on an intravenous pyelography for urolithiasis patient does not meaningfully affect the progress of contrast media.

• Imaging Science in Dentistry
• /
• v.47 no.3
• /
• pp.165-174
• /
• 2017

Purpose: This study was performed to investigate the influence of object shape and distance from the center of the image on the volumetric accuracy of cone-beam computed tomography (CBCT) scans, according to different parameters of tube voltage and current. Materials and Methods: Four geometric objects(cylinder, cube, pyramid, and hexagon) with predefined dimensions were fabricated. The objects consisted of Teflon-perfluoroalkoxy embedded in a hydrocolloid matrix (Dupli-Coe-Loid TM; GC America Inc., Alsip, IL, USA), encased in an acrylic resin cylinder assembly. An Alphard Vega Dental CT system (Asahi Roentgen Ind. Co., Ltd, Kyoto, Japan) was used to acquire CBCT images. OnDemand 3D (CyberMed Inc., Seoul, Korea) software was used for object segmentation and image analysis. The accuracy was expressed by the volume error (VE). The VE was calculated under 3 different exposure settings. The measured volumes of the objects were compared to the true volumes for statistical analysis. Results: The mean VE ranged from -4.47% to 2.35%. There was no significant relationship between an object's shape and the VE. A significant correlation was found between the distance of the object to the center of the image and the VE. Tube voltage affected the volume measurements and the VE, but tube current did not. Conclusion: The evaluated CBCT device provided satisfactory volume measurements. To assess volume measurements, it might be sufficient to use serial scans with a high resolution, but a low dose. This information may provide useful guidance for assessing volume measurements.

• Korean Journal of Remote Sensing
• /
• v.39 no.3
• /
• pp.283-295
• /
• 2023

With the development of sensor and satellite technology, numerous high-resolution and multi-spectral satellite images have been available. Due to their wavelength-dependent reflection, transmission, and scattering characteristics, multi-spectral satellite images can provide complementary information for earth observation. In particular, the short-wave infrared (SWIR) band can penetrate certain types of atmospheric aerosols from the benefit of the reduced Rayleigh scattering effect, which allows for a clearer view and more detailed information to be captured from hazed surfaces compared to the visible band. In this study, we proposed a multi-resolution transform-based image fusion method to combine visible and SWIR satellite images. The purpose of the fusion method is to generate a single integrated image that incorporates complementary information such as detailed background information from the visible band and land cover information in the haze region from the SWIR band. For this purpose, this study applied the Laplacian pyramid-based multi-resolution transform method, which is a representative image decomposition approach for image fusion. Additionally, we modified the multiresolution fusion method by combining a haze-guided weight map based on the prior knowledge that SWIR bands contain more information in pixels from the haze region. The proposed method was validated using very high-resolution satellite images from Worldview-3, containing multi-spectral visible and SWIR bands. The experimental data including hazed areas with limited visibility caused by smoke from wildfires was utilized to validate the penetration properties of the proposed fusion method. Both quantitative and visual evaluations were conducted using image quality assessment indices. The results showed that the bright features from the SWIR bands in the hazed areas were successfully fused into the integrated feature maps without any loss of detailed information from the visible bands.

• Journal of the Korea Computer Graphics Society
• /
• v.14 no.3
• /
• pp.11-17
• /
• 2008

Many methods which represent complex surfaces using height map without a number of vertex have been researched. However, a single layer height map cannot present more complex objects because it has only one height value on each position. In this paper, we introduce the new approach to render more complex objects, which are not generated by single layer height map, using multi layer height map. We store height values of the scene to each texture channel by the ascending order. A pair of ordered height values composes a geometry block and we use this property. For accurate ray search, we store the highest value in odd channels and the lowest value in even channels to generate quad tree height map. Our ray search algorithm shows accurate intersections between viewing ray and height values using quad tree height map. We solve aliasing problems on grazing angles occurred in previous methods and render the result scene on real-time.

• Journal of Korea Multimedia Society
• /
• v.12 no.7
• /
• pp.929-940
• /
• 2009

This paper presents multipurpose watermarking scheme in coutourlet transform domain for copyright protection, authentication and transform detection. Since contourlet transform can detect more multi direction edge and smooth contour than wavelet transform, the proposed scheme embeds multi watermarks in contourlet domain based on 4-level Laplacian pyramid and 2-level directional filter bank. In the first stage of the robust watermarking scheme for copyright protection, we generates the sequence of circle patterns according to watermark bits and projects these patterns into the average of magnitude coefficients of high frequency directional subbands. Then the watermark bit is embedded into variance distribution of the projected magnitude coefficients. In the second stage that is the semi-fragile watermarking scheme for authentication and transform detection, we embed the binary watermark image in the low frequency subband of higher level by using adaptive quantization modulation scheme. From the evaluation experiment using Checkmark 2.1, we verified that the proposed scheme is superior to the conventional scheme in a view of the robustness and the invisibility.