• 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 Microbiology and Biotechnology
• /
• v.31 no.4
• /
• pp.510-519
• /
• 2021

The pathological impact of haze upon the phyllosphere microbiota awaits investigation. A moderate degree of haze environment and a clean control were selected in Chengdu, China. Artemisia argyi, a ubiquitously distributed and extensively applied Chinese herb, was also chosen for experiment. Total genome DNA was extracted from leaf samples, and for metagenome sequencing, an Illumina HiSeq 2500 platform was applied. The results showed that the gene numbers of phyllosphere microbiota derived from haze leaves were lower than those of the clean control. The phyllosphere microbiota derived from both haze and clean groups shared the same top ten phyla; the abundances of Proteobacteria, Actinomycetes and Anorthococcuso of the haze group were substantially increased, while Ascomycetes and Basidiomycetes decreased. At the genus level, the abundances of Nocardia, Paracoccus, Marmoricola and Knoelia from haze leaves were markedly increased, while the yeasts were statistically decreased. KEGG retrieval demonstrated that the functional genes were most annotated to metabolism. An interesting find of this work is that the phyllosphere microbiota responsible for the synthesis of primary and secondary metabolites in A. argyi were significantly increased under a haze environment. Relatively enriched genes annotated by eggNOG belong to replication, recombination and repair, and genes classified into the glycoside hydrolase and glycosyltransferase enzymes were significantly increased. In summary, we found that both structure and function of phyllosphere microbiota are globally impacted by haze, while primary and secondary metabolites responsible for haze tolerance were considerably increased. These results suggest an adaptive strategy of plants for tolerating and confronting haze damage.

• Applied Chemistry for Engineering
• /
• v.19 no.6
• /
• pp.685-689
• /
• 2008

The effect of filler type and dispersion time on thin film characterization is studied. Hollow silica powders which have the advantage of transmittance are selected. In this paper, the characterization of haze, transmittance, and gloss are analyzed in terms of filler concentration and coating thickness, respectively. As a result, the value of haze and gloss increased with increasing of filler concentration and coating thickness, respectively. In addition, the variation of transmittance by filler concentration and coating thickness is not occurred. In conclusion, it is assumed that hollow silica powders as a filler have an advantage of concentration increment and transmittance.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.07a
• /
• pp.241-243
• /
• 2010

Previous approaches to the 2D to 3D conversion problem require heavy computation or considerable amount of user input. In this paper, we propose a rather simple method in estimating the depth map from a single image using a monocular depth cue: haze. Using the haze imaging model, we obtain the distance information and estimate a reliable depth map from a single scenery image. Using the depth map, we also suggest an algorithm that converts the single image to 3D stereoscopic images. We determine a disparity value for each pixel from the original 'left' image and generate a corresponding 'right' image. Results show that the algorithm gives well refined depth maps despite the simplicity of the approach.

• 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 the Korean Applied Science and Technology
• /
• v.26 no.1
• /
• pp.93-101
• /
• 2009

Antistatic acrylic resin is made from n-butyl methacrylate, methyl methacrylate, dimethyl amino ethyl methacrylate(DMAEMA), 2-ethyl hexyl methacrylate, hydroxy ethyl methacrylate, 2,2'-azobis iso-butyronitrile by synthesis. To achieve a lowest surface resistance of antistatic acrylic resin was applied to a variety of synthesis processes. The acrylic resin has been determined from the value of surface resistance and -then the antistatic acrylic resin including dimethyl amino ethyl methacrylate of the 10%, 20% and 30% is synthesized. Finally, dimethyl sulfate(DMS) on a variety of weight ratios is added to antistatic acrylic resin. When DMAEMA / DMS weight ratio is 1/1, antistatic acrylic resin isn't haze the lower the surface resistance. Compared to the traditional antistatic agent, all aspects of the physical properties is outstanding.

• Applied Chemistry for Engineering
• /
• v.4 no.3
• /
• pp.616-626
• /
• 1993

Polyurethane(PU) have an excellent flexibility and toughness so that it has been widely used as an elastomer. PMMA was blended with PU to improve the impact property. Five types of PU, having different molecular weight and different polyol types, were prepared and blended with PMMA in order to investigate the effect of molecular weight and polyol type of PU on property of PU-PMMA blend. Tensile strength of PU-PMMA blend was determined by Inston. Differential Scanning Calorymetry(DSC) and Scanning. Elctron Microscopy(SEM) were used to observe morphology change and glass transition temperature changes of PU-PMMA blends. Transparency of PU-PMMA blends was determined by haze meter. But, owing to intrinsic incompatability of PU-PMMA, Low impact strength of PMMA wasn't improved through PU-PMMA blend. therefore, polyurethane dimethacrylate(PUD), having similiar chemical structure to PU and two vinyl group at both ends, was prepared and reacted with methyl methacrylate(MMA) to form crosslinked copolymer Mechanical property of this crosslinked polymer, such as impact strength and transparency, was investigated by Instron, Izod type (Cantilever beam) impact tester and haze meter. Results of these measurements showed that crosslinked copolymer of PUD-MMA was better impact resistance than PMMA and maintained similar transparency to PMMA.

• 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.