Abstract:In practical fuzzy applications, such as image processing, the utilization of precise models in hardware may not be the most efficient approach due to increased energy consumption and chip resource allocation. In a fuzzy system, an approximate implementation of min-max blocks offers an efficient solution with minimal accuracy compromise. Nevertheless, recent designs predominantly employ non-commercialized technologies for the fundamental fuzzy blocks. This paper introduces a novel hardware solution for approximate fuzzy image edge detection using the well-established Independent Gate (IG) Fin FieldEffect Transistor (FinFET) technology. The proposed hardware leverages two inference rules to identify edge pixels effectively. The fuzzy inference engine is implemented at the circuit level using 24 FinFETs, while the defuzzifier section incorporates four FinFETs with a configurable thresholding structure for optimal performance. Our circuit-level simulations reveal a remarkable 71% reduction in energy consumption compared to previous designs. The edge detection results are compared at the system level with the MATLAB Sobel edge detector. The proposed approximate hardware consistently matches the Sobel edge detection outcomes, exhibiting a 15% average improvements in data loss rate than other approximate structures. A Figure of Merit (FoM) is introduced to provide a comprehensive evaluation, considering both circuit and system-level metrics. The proposed FinFET-based hardware outperforms other approximate and even exact fuzzy edge detection hardware designs, boasting a 1.8 times higher FoM. This design paradigm exemplifies a promising direction toward compact and energy-efficient on-chip hardware implementations of real-world fuzzy systems.

Design Principle of an Analogue/digital Hybrid Construction Basedon Functioning-Fuzzy-subset Inference

Hardware Implementation of Fast Huffman Coding Based on Different Sorting Methods

Compiling Fuzzy Logic Control Rules to Hardware Implementations

A VLSI Design and Implementation for a Real-Time Approximate Reasoning

Hybrid Reasoning Framework Integrating of Rule-Based and Case-Based Reasoning

High Speed Fuzzy Logic Microprocessor and Its Applications in a Real-Time Hardware Expert System

Cmos Design of Analog Fuzzy System

A CMOS Current-Mode High Speed Fuzzy Logic Microprocessor for a Real-Time Expert System.

A Hardware Realization Framework for Fuzzy Inference System Optimization

Hybrid Model of Intelligent Control System with Fuzzy Logical Interface

Digital-analog mixed signal processing system for imprecise computation

FPGA Implementation of Feature Detection Algorithm Based on High Level Synthesis

Design of Adder Based on Abacus Algorithm

An AND-LUT Based Hybrid FPGA Architecture

Hardware Thread Accelerating Method Based on CPU/FPGA Hybrid Architecture

HIGH-SPEED ECL CIRCUIT FOR LOOKING UP FUZZY CONTROL TABLE BASED ON MULTIVALUED LOGIC

CMOS CIRCUIT DESIGN OF A TAKAGI-SUGENO FUZZY LOGIC CONTROLLER

A Novel Hybrid FPGA Architecture

Neuro-fuzzy System with High-Speed Low-Power Analog Blocks

Implementation of intra prediction in H.264 based on a novel design methodology

A Novel Hardware Solution for Efficient Approximate Fuzzy Image Edge Detection