• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.9
• /
• pp.2882-2903
• /
• 2022

Image super-resolution (SR) processing is of great value in the fields of digital image processing, intelligent security, film and television production and so on. This paper proposed a densely connected deep learning network based on cascade architecture, which can be used to solve the problem of super-resolution in the field of image quality enhancement. We proposed a more efficient residual scaling dense block (RSDB) and the multi-channel cascade architecture to realize more efficient feature reuse. Also we proposed a hybrid loss function based on L₁ error and L_∞ error to achieve better L_∞ error performance. The experimental results show that the overall performance of the network is effectively improved on cascade architecture and residual scaling. Compared with the residual dense net (RDN), the PSNR / SSIM of the new method is improved by 2.24% / 1.44% respectively, and the L_∞ performance is improved by 3.64%. It shows that the cascade connection and residual scaling method can effectively realize feature reuse, improving the residual convergence speed and learning efficiency of our network. The L_∞ performance is improved by 11.09% with only a minimal loses of 1.14% / 0.60% on PSNR / SSIM performance after adopting the new loss function. That is to say, the L_∞ performance can be improved greatly on the new loss function with a minor loss of PSNR / SSIM performance, which is of great value in L_∞ error sensitive tasks.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.39 no.4
• /
• pp.104-111
• /
• 2002

In this paper, we proposed the architecture of the decoder which implements the multi-view images decoding algorithm. The study of the hardware structure of the multi-view image processing has not been accomplished. The proposed multi-view images decoder operates in a three stage pipelined manner and extracts the depth of the pixels of the decoded image every clock. The multi-view images decoder consists of three modules, Node selector which transfers the value of the nodes repeatedly and Depth Extractor which extracts the depth of each pixel from the four values of the nodes and Affine transformer which generates the projecting position on the image plane from the values of the pixels and the specified viewpoint. The proposed architecture is designed and simulated by the Max+plus II design tool and the operating frequency is 30MHz. The image can be constructed in a real time by the decoder with the proposed architecture.

• Journal of the Regional Association of Architectural Institute of Korea
• /
• v.9 no.3
• /
• pp.1-8
• /
• 2007

Heidegger's speculation on "Disclosure" makes it possible to know on image expression fundamentally. "Disclosure" means that the 'earth' and the 'world' are not fixed into one, but harmonize dynamically with each other under continuous mutual tension and mutual support. When they are well-weaved with each other, the two ways are collectively defined as "disclosure Image". In regard to the technique, the study analyzed 'allegory' and 'reactivation'. The former refers to showing image potentially, while the latter means embodying and revealing image apparently. Le Corbusier's late Architecture provides a simultaneous reactivation of various images through continuous cross-weaving rather than being inclined to either the earth or the world.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.10
• /
• pp.2043-2052
• /
• 1994

This paper proposes a new parallel ASIC architecture for real-time image processing to reduce inter-processing element (inter-PE) communication overhead, called a Sliding Memory Plane (SliM) Image Processor. The Slim Image Processor consists of $3\times3$ processing elements (PEs) connected by a mesh topology. With easy scalability due to the topology. a set of SliM Image Processors can form a mesh-connected SIMD parallel architecture. called the SliM Array Processor. The idea of sliding means that all pixels are slided into all neighboring PEs without interrupting PEs and without a coprocessor or a DMA controller. Since the inter-PE communication and computation occur simultaneously. the inter-PE communication overhead, significant disadvantage of existing machines greatly diminishes. Two I/O planes provide a buffering capability and reduce the date I/O overhead. In addition, using the by-passing path provides eight-way connectivity even with four links. with these salient features. SliM shows a significant performance improvement. This paper presents architectures of a PE and the SliM Image Processor, and describes the design of an instruction set.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.32 no.1
• /
• pp.57-68
• /
• 2004

The purpose of this paper is to discover the Image and Visual Preference of the Environmental Sculpture in Urban Streetscape. For this, the analysis was preformed by the data obtained from questionnaires and from the photos of the environmental sculptures scene. The landscape image was analyzed by the factor analysis algorithm. The level of visual preference was measured by a slide simulation test, and this data was analyzed by multiple regression. The results of this study can be summarized as follows: The visual preference of the environmental sculpture has been evaluated to average 4.03 on a scale of 7 Landscape slides No. 11 and No.5 were ranked more highly for visual preference. Factors formulating the landscape image were found to be 'beauty', 'orderliness', 'emotion', and 'formation'. By using the control method for the number of factors, T.V., were obtained as 63.0%. For all experimental landscape slides, the factor of orderliness was found to be the main factor determining the visual preference. The 4 factors for visual preference were analyzed by regression as follows: Visual Preference = 3.996 + 0.341(FS1) + 0.595(FS2) + 0.222(FS3) + 0.011(FS4), R-Square = 0.520.

• Proceedings of the Korean Institute of Landscape Architecture Conference
• /
• 1999.03a
• /
• pp.26-27
• /
• 1999

• Proceedings of the Korean Institute of Landscape Architecture Conference
• /
• 1999.03a
• /
• pp.20-21
• /
• 1999

• Journal of the Korean Institute of Landscape Architecture
• /
• v.35 no.4
• /
• pp.81-89
• /
• 2007

Urban landscape represents a city's internal identity as well as external form. Therefore, urban landscape and image planning has to involve marketing strategies to deliver the economic growth and regional revitalization through recreating the environmental and cultural value of the city. This study intends to develop practical city image management programs as city marketing tools which consider opinions of local authorities, related experts, and residents. To achieve this purpose conjoint analysis was applied on three components which symbolically reflects features of assets and resources in Anseong. This study can be summarized as follows: 1. The first consideration for Lake Geumgwang is to develop farm experience Programs on waterside fields. 2. The rot consideration for Bauduki Festival is to make efforts for public relations(PR) to be able to merchandise festival-related goods and develop unique products. 3. The first consideration for Anseong Grapes is to build leisure-oriented theme parks for local farm products. This study is meaningful to develop the efficient and quantitative management method of city image for marketing. The integrated approach to the management programs to reflect opinions of local policies, experts and citizens is expected to become a useful foundation in studying city images and marketing.

• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.1137-1140
• /
• 1998

In this paper, we propose the new hardware architecture which implements the stereo matching algorithm using the dynamic programming method. The dynamic programming method is used in finding the corresponding pixels between the left image and the right image. The proposed MOEPE(Merged Odd-Even PE) architecture operates in the systolic manner and finds the disparities from the intensities of the pixels on the epipolar line. The number of PEs used in the MOEPE architecture is the number of the range constraint, which reduced the number of the necessary PEs dramatically compared to the traditional method which uses the PEs with the number of pixels on the epipolar line. For the normal method by 25 times. The proposed architecture is modeled with the VHDL code and simulated by the SYNOPSYS tool.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2005.07b
• /
• pp.850-853
• /
• 2005

This paper presents issues of display hardware architecture, relating to memory, display driver IC architecture, and chip-to-chip interface. To achieve a low power and low cost mobile phone, not only the display architecture must be carefully selected, but also the driver-ICs optimized to accommodate the different modes of operation found in typical handheld devices. The technique of forming a photo sensor in each pixel using TFT and display module architecture are developed to add multi functions in display such as fingerprint recognition, image scanning, and integrated touch screen. Detailed architectures of IC partitioning, high-speed serial interface, D/A converter, and multi functions such as fingerprint recognition and image scanning using photo sensors are important to a power optimized system.