RETRACTED ARTICLE: A novel patch selection technique in ANN B-Spline Bayesian hyperprior interpolation VLSI architecture using fuzzy logic for highspeed satellite image processing

Autor: C. Vennila, K. Chitra
Rok vydání: 2020
Předmět:
Zdroj: Journal of Ambient Intelligence and Humanized Computing. 12:6491-6504
ISSN: 1868-5145
1868-5137
DOI: 10.1007/s12652-020-02264-9
Popis: During recent days various research are going on Satellite image processing including remote sensing, agriculture, disaster management and various other fields. Denoising and interpolation are two major tasks in Image processing. These tasks assist screening and interpreting many scientific images such as medical images and satellite images. The proposed work named as “fuzzy logic-based Patch Selection in ANN B-Spline Bayesian hyperprior interpolation VLSI architecture” (FLABBHI) is contrived in the indent of designing a new VLSI architecture for faster satellite image denoising and interpolation. Non-native image sampling process for denoising is performed using fuzzy logic based patch selection and artificial neural network-based B-Spline Bayesian hyperprior interpolation (ABBHI) procedure is used for interpolation process. Fuzzy logic is used to improve the execution speed in the process of selecting patch images from a collection of received images to construct the image more similar toground-truth image. For experimentation, FLABBHI is designed for the target hardware Kintex, used to construct and evaluate the proposed method. Xilinx ISE is used to measure the performance metrics of the target FPGA. The interpolation is performed by ABBHI method to improve the quality of the constructed image during the interpolation process. A real-time mode is included in the proposed FLABBHI to buffer and process incoming images from a satellite lively. This real-time mode is equipped to run with the incremental rendering process which is used to get a quick scanning in the beginning and more accurate reconstructed image over the time.
Databáze: OpenAIRE