• Journal of Information Processing Systems
• /
• v.14 no.6
• /
• pp.1445-1456
• /
• 2018

Environment perception and three-dimensional (3D) reconstruction tasks are used to provide unmanned ground vehicle (UGV) with driving awareness interfaces. The speed of obstacle segmentation and surrounding terrain reconstruction crucially influences decision making in UGVs. To increase the processing speed of environment information analysis, we develop a CPU-GPU hybrid system of automatic environment perception and 3D terrain reconstruction based on the integration of multiple sensors. The system consists of three functional modules, namely, multi-sensor data collection and pre-processing, environment perception, and 3D reconstruction. To integrate individual datasets collected from different sensors, the pre-processing function registers the sensed LiDAR (light detection and ranging) point clouds, video sequences, and motion information into a global terrain model after filtering redundant and noise data according to the redundancy removal principle. In the environment perception module, the registered discrete points are clustered into ground surface and individual objects by using a ground segmentation method and a connected component labeling algorithm. The estimated ground surface and non-ground objects indicate the terrain to be traversed and obstacles in the environment, thus creating driving awareness. The 3D reconstruction module calibrates the projection matrix between the mounted LiDAR and cameras to map the local point clouds onto the captured video images. Texture meshes and color particle models are used to reconstruct the ground surface and objects of the 3D terrain model, respectively. To accelerate the proposed system, we apply the GPU parallel computation method to implement the applied computer graphics and image processing algorithms in parallel.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.139-141
• /
• 2022

As interest in digital video media and intelligent systems increases rapidly, technologies using video information are being combined and used in various fields such as security and artificial intelligence. Impulse noise generated during digital image processing degrades the image quality of the image and reduces the reliability of information, so it is necessary to remove it through a filter. There are SMF, AWMF, and MDBUTMF as well-known antecedent methods, but they all have limitations in achieving seamless filtering in environments with large loss of information on valid pixels due to problems with the algorithm itself. Therefore, this paper designs a median filter algorithm that applies weights reflecting the reliability of the information by searching for the nearest effective pixels present within the mask. For performance evaluation, this algorithm and the preceding algorithm were compared and analyzed using PSNR and enlarged images.

• Journal of Broadcast Engineering
• /
• v.11 no.4 s.33
• /
• pp.471-482
• /
• 2006

In this paper, we propose a new scheme to generate multi-view images utilizing depth map decomposition and adaptive edge smoothing. After carrying out smooth filtering based on an adaptive window size to regions of edges in the depth map, we decompose the smoothed depth map into four types of images: regular mesh, object boundary, feature point, and number-of-layer images. Then, we generate 3-D scenes from the decomposed images using a 3-D mesh triangulation technique. Finally, we extract multi-view images from the reconstructed 3-D scenes by changing the position of a virtual camera in the 3-D space. Experimental results show that our scheme generates multi-view images successfully by minimizing a rubber-sheet problem using edge smoothing, and renders consecutive 3-D scenes in real time through information decomposition of depth maps. In addition, the proposed scheme can be used for 3-D applications that need the depth information, such as depth keying, since we can preserve the depth data unlike the previous unsymmetric filtering method.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.47 no.5
• /
• pp.43-51
• /
• 2010

Attenuation coefficients of medical ultrasound not only reflect the pathological information of tissues scanned but also provide the quantitative information to compensate the decay of backscattered signals for other medical ultrasound parameters. Based on the frequency-selective attenuation property of human tissues, attenuation estimation methods in spectral domain have difficulties for real-time implementation due to the complexicity while estimation methods in time domain do not achieve the compensation for the diffraction effect effectively. In this paper, we propose the modified VSA method, which compensates the diffraction with reference phantom in time domain, using adaptive bandpass filters with decreasing center frequencies along depths. The adaptive bandpass filtering technique minimizes the distortion of relative echogenicity of wideband transmit pulses and maximizes the signal-to-noise ratio due to the random scattering, especially at deeper depths. Since the filtering center frequencies change according to the accumulated attenuation, the proposed algorithm improves estimation accuracy and precision comparing to the fixed filtering method. Computer simulation and experimental results using tissue-mimicking phantoms demonstrate that the distortion of relative echogenicity is decreased at deeper depths, and the accuracy of attenuation estimation is improved by 5.1% and the standard deviation is decreased by 46.9% for the entire scan depth.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.4
• /
• pp.479-488
• /
• 2002

In recent years the amount of digital video used has risen dramatically to keep pace with the increasing use of the Internet and consequently an automated method is needed for indexing digital video databases. Textual information, both superimposed and embedded scene texts, appearing in a digital video can be a crucial clue for helping the video indexing. In this paper, a new method is presented to extract both superimposed and embedded scene texts in a freeze-frame of news video. The algorithm is summarized in the following three steps. For the first step, a color image is converted into a gray-level image and applies contrast stretching to enhance the contrast of the input image. Then, a modified local adaptive thresholding is applied to the contrast-stretched image. The second step is divided into three processes: eliminating text-like components by applying erosion, dilation, and (OpenClose+CloseOpen)/2 morphological operations, maintaining text components using (OpenClose+CloseOpen)/2 operation with a new Geo-correction method, and subtracting two result images for eliminating false-positive components further. In the third filtering step, the characteristics of each component such as the ratio of the number of pixels in each candidate component to the number of its boundary pixels and the ratio of the minor to the major axis of each bounding box are used. Acceptable results have been obtained using the proposed method on 300 news images with a recognition rate of 93.6%. Also, my method indicates a good performance on all the various kinds of images by adjusting the size of the structuring element.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.05a
• /
• pp.527-530
• /
• 2008

The watermarking method was suggested to protect the right of the copyright and the right of property, data loss is yes or not, copy prevention and pursuit about the brainchild(music, image, video and so on) in the early 1990. In this paper, we suggest the watermarking algorithm against the geometrical transformation of the medical images applied the Watershed algorithm. The robust watermarking methodology is being suggested using the several algorithm against the attacks such as compression, filtering, noise and geometrical transformation of the medical image.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.2 s.95
• /
• pp.39-49
• /
• 1999

This paper describes the development of a vision-aided position and orientation recognition system for automatically adjusting electronic tuners which control the waveform by rotating variable resisters. The position and orientation recognition system estimates the center and the angle of the tuner grooves so that the main controller may correct the difference from the ideal position and thereby manipulate the variable resisters automatically. In this paper a robust algorithm is suggested which estimates the center and the angle of the tuner grooves fast and precisly from the source image with lighting variance and video noise. In the algorithm morphological filtering, 8-chain coding, and invariant moments are sequentially used to figure out image segments concerned. The performance of the proposed system was evaluated using a set of real specimens. The results indicate the system works well enough to be used practically in real manufacturing lines. If the system adopts a high speed frame grabber which enables real time image processing, it can also be applied to positioning of robot manipulators as well as automated PCB adjusters.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.5 no.11
• /
• pp.2016-2034
• /
• 2011

Ubiquitous Healthcare (u-Healthcare) is the intelligent delivery of healthcare services to users anytime and anywhere. To provide robust healthcare services, recognition of patient daily life activities is required. Context information in combination with user real-time daily life activities can help in the provision of more personalized services, service suggestions, and changes in system behavior based on user profile for better healthcare services. In this paper, we focus on the intelligent manipulation of activities using the Context-aware Activity Manipulation Engine (CAME) core of the Human Activity Recognition Engine (HARE). The activities are recognized using video-based, wearable sensor-based, and location-based activity recognition engines. An ontology-based activity fusion with subject profile information for personalized system response is achieved. CAME receives real-time low level activities and infers higher level activities, situation analysis, personalized service suggestions, and makes appropriate decisions. A two-phase filtering technique is applied for intelligent processing of information (represented in ontology) and making appropriate decisions based on rules (incorporating expert knowledge). The experimental results for intelligent processing of activity information showed relatively better accuracy. Moreover, CAME is extended with activity filters and T-Box inference that resulted in better accuracy and response time in comparison to initial results of CAME.

• Journal of IKEEE
• /
• v.2 no.1 s.2
• /
• pp.96-100
• /
• 1998

Due to expected wide spread use of DCT based image/video coding standard, it is advantageous to process data directly in the DCT domain rather than decoding the source back to the spatial domain. The block processing algorithm provides a parallel processing method since multiple input data are processed in the block filter structure. Hence a fast implementation of the algorithm is well suited. In this paper, we propose the JPEG browse by Block Transform Domain Filtering(BTDF) using subband filter banks. Instead of decompressing the entire image to retrieve at full resolution from compressed format, a user can select the level of expansion required$(2^N{\times}2^N)$. Also this approach reduces the computer cpu time by reducing the number of multiplication through BTDF in the filter banks.

• Journal of Broadcast Engineering
• /
• v.16 no.5
• /
• pp.782-792
• /
• 2011

In this paper, we address a new fast DCT-II/DFT/HWT hybrid transform architecture for digital video and fusion mobile handsets based on Jacket-like sparse matrix decomposition. This fast hybrid architecture is consist of source coding standard as MPEG-4, JPEG 2000 and digital filtering discrete Fourier transform, and has two operations: one is block-wise inverse Jacket matrix (BIJM) for DCT-II, and the other is element-wise inverse Jacket matrix (EIJM) for DFT/HWT. They have similar recursive computational fashion, which mean all of them can be decomposed to Kronecker products of an identity Hadamard matrix and a successively lower order sparse matrix. Based on this trait, we can develop a single chip of fast hybrid algorithm architecture for intelligent mobile handsets.