• Journal of the Korea Computer Graphics Society
• /
• v.28 no.3
• /
• pp.45-54
• /
• 2022

Image depth estimation is a technology that is the basis of various image analysis. As analysis methods using deep learning models emerge, studies using deep learning in image depth estimation are being actively conducted. Currently, most deep learning-based depth estimation models are being trained with clean and ideal images. However, due to the lack of data on adverse conditions such as haze or fog, the depth estimation may not work well in such an environment. It is hard to sufficiently secure an image in these environments, and in particular, obtaining non-homogeneous haze data is a very difficult problem. In order to solve this problem, in this study, we propose a method of synthesizing non-homogeneous haze images and a learning method for a monocular depth estimation deep learning model using this method. Considering that haze mainly occurs outdoors, datasets mainly containing outdoor images are constructed. Experiment results show that the model with the proposed method is good at estimating depth in both synthesized and real haze data.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.8
• /
• pp.45-52
• /
• 2017

This paper presents a high-performance software-hardware dehazing system based on a dedicated hardware accelerator for the haze removal. In the proposed system, the dedicated hardware accelerator performs the dark-channel-prior-based dehazing process, and the software performs the other control processes. For this purpose, the dehazing process is realized as an OpenCL kernel by finding the inherent parallelism in the algorithm and is synthesized into a hardware by employing a high-level-synthesis technique. The proposed system executes the dehazing process much faster than the previous software-only dehazing system: the performance improvement is up to 96.3% in terms of the execution time.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.12
• /
• pp.571-576
• /
• 2015

Image dehazing has been extensively studied, but the performance evaluation method for dehazing techniques has not attracted significant interest. This paper surveys many existing performance evaluation methods of image dehazing. In order to analyze the reliability of the evaluation methods, synthetic hazy images are first reconstructed using the ground-truth color and depth image pairs, and the dehazed images are then compared with the original haze-free images. Meanwhile we also evaluate dehazing algorithms not by the dehazed images' quality but by the performance of computer vision algorithms before/after applying image dehazing. All the aforementioned evaluation methods are analyzed and compared, and research direction for improving the existing methods is discussed.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.11
• /
• pp.31-37
• /
• 2018

This paper proposes an effective single-image haze-removal algorithm with low complexity by using a dual dark channel prior (DCP) and an adaptive brightness correction technique. The dark channel of a small patch preserves the edge information of the image, but is sensitive to noise and local brightness variations. On the other hand, the dark channel of a large patch is advantageous in estimation of the exact haze value, but halo effects from block effects deteriorate haze-removal performance. In order to solve this problem, the proposed algorithm builds a dual DCP as a combination of dark channels from patches with different sizes, and this meets low-memory and low-complexity requirements, while the conventional method uses a matting technique, which requires a large amount of memory and heavy computations. Moreover, an adaptive brightness correction technique that is applied to the recovered image preserves the objects in the image more clearly. Experimental results for various hazy images demonstrate that the proposed algorithm removes haze effectively, while requiring much fewer computations and less memory than conventional methods.

• Journal of Broadcast Engineering
• /
• v.19 no.5
• /
• pp.656-664
• /
• 2014

In the case of a haze image, transferring the information of the original image is difficult as the contrast leans toward bright regions. Thus, dehazing algorithms have become an important area of study. Normally, since it is hard to obtain a haze-free image, the output image is qualitatively analyzed to test the performance of an algorithm. However, this paper proposes a quantitative error comparison based on reproducing the haze image using a disparity map. In addition, a Hidden Random Markov Model and EM algorithm are used to remove any block artifacts. The performance of the proposed algorithm is confirmed using a variety of synthetic and natural images.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.6
• /
• pp.819-824
• /
• 2021

As technologies related to image processing such as autonomous driving and CCTV develop, fog removal algorithms using a single image are being studied to improve the problem of image distortion. As a method of predicting fog density, there is a method of estimating the depth of an image by generating a depth map, and various fog features may be used as training data of the depth map. In addition, it is essential to implement a hardware capable of processing high-definition images in real time in order to apply the fog removal algorithm to actual technologies. In this paper, we implement NLCV (Normalize Local Coefficient of Variation), a feature of fog based on coefficient of variation, in hardware. The proposed hardware is an FPGA implementation of Xilinx's xczu7ev-2ffvc1156 as a target device. As a result of synthesis through the Vivado program, it has a maximum operating frequency of 479.616MHz and shows that real-time processing is possible in 4K UHD environment.

• Journal of Plant Biotechnology
• /
• v.37 no.1
• /
• pp.89-95
• /
• 2010

Dihydroflavonol 4-reductase (DFR) is a key enzyme of the flavonoid biosynthesis pathway which catalyses the NADPH-dependent reduction of 2R,3R-trans-dihydroflavonols to leucoanthocyanidins. In this study we describe cloning and expression of the genes encoding the flavonoid-biosynthetic enzyme DFR in Gypsophila paniculata L. Inspection of the 1279 bp long sequence revealed an open reading frame 1063 bp, including a 36 bp 5' leader region and 181 bp 3' untranslated region. Comparison of the coding region of this DFR cDNA sequence including the sequences of Arabidopsis thaliana, Citrus sinensis, Dianthus caryophyllus, Ipomoea batatas, Matthiola incana, Nierembergia sp, Petunia hybrida, Solanum tuberosum, Vitis vinifera reveals an identity higher than 69% at the nucleotide level. The function of this nucleotide sequences was verified by comparison with amino acid sequences of the amino-terminus and tryptic peptides from purified plant enzyme, by northern blotting with mRNA from wild type and mutant plants, by in vitro expression yielding and enzymatically active reductase, as indicated by the small leucopelargonidin peak. Genomic southern blot analysis showed the presence of only one gene for DFR in Gypsophila paniculata.

• Journal of IKEEE
• /
• v.26 no.3
• /
• pp.401-407
• /
• 2022

Recently, image processing technology for autonomous driving by recognizing objects and lanes through camera images to realize autonomous vehicles is being studied. Haze reduces the visibility of images captured by the camera and causes malfunctions of autonomous vehicles. To solve this, it is necessary to apply the haze removal function that can be processed in real time to the camera. Therefore, in this paper, the fog removal method of Sim with excellent performance is implemented with hardware capable of real-time processing. The proposed hardware was designed using Verilog HDL, and FPGA was implemented by setting Xilinx's xc7z045-2ffg900 as the target device. As a result of logic synthesis using Xilinx Vivado program, it has a maximum operating frequency of 276.932MHz and a maximum processing speed of 31.279fps in a 4K (4096×2160) high-resolution environment, thus satisfying the real-time processing standard.

• Journal of Plant Biotechnology
• /
• v.33 no.2
• /
• pp.85-91
• /
• 2006

Flavanone 3$\beta$-hydroxylase (FHT) is an enzyme acting in the central part of the flavonoid biosynthesis pathway. FHT catalyses the hydroxylation of flavanone to dihydroflavonols in the anthocyanin pathway. In this paper we describe the cloning and expression of the genes encoding the flavonoid-biosynthetic enzyme FHT in Gypsophila paniculata L. A heterologous cDHA probe from Dianthus cavophyllus was used to isolate FHT-encoding cDHA clones from Gypsophila paniculata L.. Inspection of the 1471 bp long sequence revealed an open reading frame 1047 bp, including a 190 bp 5' leader region and 288 bp 3' untranslated region. Comparison of the coding region of this FHT cDHA sequence including the sequences of Arabidopsis thaliana, Citrus sinensis, Dianthus caryophyllus, Ipomoea batatas, Matthiola incana, Nierembergia sp, Petunia hybrida, Solanum tuberosum, Vitis vinifera reveals a identity higher than 69% at the nucleotide level. The function of this nucleotide sequences were verified by comparison with amino acid sequences of the amino-terminus and tryptic peptides from purified plant enzyme, by northern blotting with mRHA from wild type and mutant plants, by in vitro expression yielding and enzymatically active hydroxylase, as indicated by the small dihydrokaempferol peak. Genomic southern blot analysis showed the presence of only one gene for FHT in Gypsophila paniculata.

• 한국신재생에너지학회:학술대회논문집
• /
• 2010.06a
• /
• pp.185.1-185.1
• /
• 2010

원격탐사(remote sensing)란 관측 대상과의 접촉 없이 멀리서 정보를 얻어내는 기술을 말한다. 기상관측분야에는 이미 소다(SODAR) 장비가 폭넓게 사용되거 왔으나 최근 풍력자원평가(wind resource assessment)를 위한 풍황측정에 SODAR와 더불어 라이다(LIDAR)가 적극적으로 활용되기 시작하고 있다. 참고로 SODAR(SOnic Detection And Ranging)는 수직 및 동서 남북 방향으로 음파를 발생시키고 대기유동에 의해 산란 반사된 에코를 수신하여 진동수 변화와 반사에코 강도를 측정하여 각 방향의 에코자료를 벡터 합성함으로써 풍향 및 풍속을 산출하는 원리이다. 반면 LIDAR(Light Detection And Ranging)는 비교적 최근에 풍황측정 용도로 개발된 레이저 탐지에 바탕을 둔 원거리 센서로, 공기입자(먼지, 수증기, 구름, 안개, 오염물질 등)에 의해 산란된 레이저 발산의 도플러 쉬프트(Doppler shift)를 이용하여 풍향 및 풍속을 측정하는 원격탐사 장비이다. 풍력자원평가 측면에서 라이다는 그 정확도가 IEC61400-12에 의거한 풍황탑(met-mast) 측정자료 다수와의 비교검증 실측평가(Albers et al., 2009)를 통하여 입증된 바 있다. 한편 한국에너지기술연구원에서 운용 중인 라이다 시스템은 그림 1의 우측 그림과 같이 1초에 $360^{\circ}$를 스캔하여 50지점에서 반사되는 레이저를 스펙트럼으로 측정하되 설정된 관측높이에서 풍속은 샘플링 부피(sampling volume)의 평균값으로 정의된다. 그런데 샘플링 부피는 설정된 관측높이로부터 상하 12.5m, 총 25m의 높이구간에서 관측한 스펙트럼의 평균값을 그 중앙지점에서의 풍속으로 환산하는 알고리듬(algorithm)을 채택하고 있다. 따라서 비선형적으로 변화하는 풍속연직분포 관측 시 풍속환산 알고리듬에 의한 측정오차가 개입될 가능성이 존재하는 것이다. 이에 본 연구에서는 라이다에 의한 풍속연직분포 측정 시 샘플링 부피의 구간 평균화 과정에서 발생하는 불확도(uncertainty)를 정량적으로 분석함으로써 라이다에 의한 풍속연직분포 관측의 불확도를 정량평가하고자 한다.