Image Processing

A Tiny CNN Architecture for Medical Face Mask Detection for Resource-Constrained Endpoints

A model of just 138KB working at 30fps, is proposed for face-mask detection on ARM Cortex-M7 microcontroller clocked at 480 Mhz and having just 496 KB of framebuffer RAM. This technique is fast, reliable, safe and helps thwart the transmission of the coronavirus.

An Efficient Analyses of the Behavior of One Dimensional Chaotic Maps using 0-1 Test and Three State Test

Strong and weak regions of Logistic Map, Logistic-Tent system, Logistic-Sine system and Tent-Sine system have been identified using using 0-1 test and three state test (3ST). This'll encourage stronger testing practices, making cryptosystems more secure.

Recent Advances in Selective Image Encryption and its Indispensability due to COVID-19

This paper discusses how COVID-19 has changed the landscape of multimedia consumption and resulted in an imperative need for better and faster selective encryption techniques and presents a survey of the most recent advances.

Rethinking Generalization in American Sign Language Prediction for Edge Devices with Extremely Low Memory Footprint

A model of just 185KB is proposed for ASL prediction on ARM Cortex-M7 microcontroller, which works at 20 fps. Interpolation augmenation is shown to be effective for improving classification generalization.

A Selective Block Cipher Image Encryption Algorithm based on Otsu-automated Canny Edge Detection, 4-D Hyperchaotic System, Improved Logistic Map and Square Wave SFC

Novel selective image encryption technique based on 4-D hyperchaos, improved 1-D Logistic map, and the square wave space filling curve (SFC) is proposed that outperforms existing schemes.

An Adaptive-Selective Image Encryption with Orthogonal Polynomials, Chaos, LU decomposition and Square-wave Shuffling

A domain-flexible and selective image encryption scheme based on genetic algorithm, non-linear chaotic map, square-wave diffusion and orthogonal polynomials transformation is proposed. Extensive experiments were used to demonstrate efficiency and practicality.

An Adaptive-Selective Image Encryption with Orthogonal Polynomials, Chaos, LU decomposition and Square-wave Shuffling

A new selective image encryption technique based on hyperchaos, random artificial image, orthogonal polynomials transformation and dynamic permutation matrix using Lower Upper (LU) Decomposition is proposed, that outperforms existing approaches.