Image Interpolation Based on Spiking Neural Network Model

Abstract

Image interpolation is used in many areas of image processing. It is seen that many techniques developed to date have been successful in both protecting edges and increasing image quality. However, these techniques generally detect edges with gradient-based linear calculations. In this study, spiking neural networks (SNNs), which are known to successfully simulate the human visual system (HVS), are used to detect edge pixels instead of the gradient. With the help of the proposed SNN-based model, the pixels marked as edges are interpolated with a 1D directional filter. For the remaining pixels, the standard bicubic interpolation technique is used. Additionally, the success of the proposed method is compared to known methods using various metrics. The experimental results show that the proposed method is more successful than the other methods.

Revolutionizing Literacy: The Life of Omar ibn Said, Written by Himself

“…A plaque in his honor mentions his autobiography and conversion to Christianity, and a mosque named after him in Fayetteville, North Carolina. For the next six decades, acknowledgments of his written documents appeared in occasional newspaper articles and religious journals, before an article written by Jameson (1925). He included Qur’anic verses from the autobiography in his article, noting the decision was necessary because they were, “an important part of the man, and help to give a narrative greater completeness as a ‘human document,’ unusual and indeed of somewhat pathetic interest” (p. 792).…”

“…I am a Christian with a limited knowledge of the Islamic religion. When speaking with a former doctoral student, Incetas (2022), a Muslim man who is literate in ancient Arabic having been a translator for the U.S. government (who knew that it would have come in handy for my research), he explained that although there are many dialects of Arabic, the text of the Qur'an is strictly controlled, to "preserve its originality." He states that early popular versions were written in Egypt, but the current translation from Saudi Arabia is now the most popular.…”

Image interpolation with spiking neural network based pixel similarity

“…Pixel similarity shows effective performance in edge detection, it is based on linear calculation of the difference between pixels. On the other hand, the SNN-based model, which detects edges in accordance with HVS proposed in 2023, appears to be quite successful [31]. However, that study presented a complex model in which calculations had to be made on all 9 pixels in the 3 × 3 window at each stage when determining the edges.…”

“…This approach has shown successful results. On the other hand, an approach proposed by İncetaş in 2023 is an significant study where edges are determined nonlinearly [31]. The author utilized the well-known spike neural network (SNN) architecture, which successfully simulates the human visual system.…”