2021 9th International Symposium on Next Generation Electronics (ISNE2021)

81 / 2020-04-12 12:37:14

Robust Real-time Super-resolution Images Reconstruction Technique

image reconstruction; SURF; adaptive normalized cnvolution

IOT and AI Techniques > Internet of Things

The technology of super-resolution reconstruction has been widely used in the intelligent transport system. This paper proposes a new super-resolution image reconstruction method with real-time robustness, which is used to improve the quality of super-resolution image reconstruction. Firstly, this method extracts the SURF feature descriptors with uniqueness and robustness from the low-resolution image, secondly, take its matching calculation, then use the least squares method to estimate motion parameters, and finally take the use of the ANC algorithm for image interpolation.The experimental result shows that the method has strong robustness and real-time and higher reconstruction quality compared with other reconstruction algorithms.