FPGA modular based implementation of a grayscale morphological camera for real-time applications

The purpose of this thesis is to develop an FPGA Modular Based Implementation of a Grayscale Morphological Camera for Real-time Applications. A key application is the development of an intelligent camera that satisfies the real-world constraints associated with autonomous vehicles: real-time performance, low power operation, and limitations on weight and size. -- This intelligent camera consists of an image sensor and a programmable logic device called an FPGA (Field Programmable Gat e Array) that serves as a reconfigurable hardware preprocessor. This preprocessing can filter out information that may be regarded as less relevant, while preserving the important structural properties of an image. Processing functions on the preprocessor include histogram generation, edge detection and mathematical morphological operators. Although computationally expensive, mathematical morphological operators are a class of powerful image processing tools for applications such as detecting small targets in cluttered environments, as might be required by an intelligent camera. -- FPGA devices have a significant advantage over serial CP Us and DSPs due to their extremely high throughput rate and ability to exploit parallelism. This research aims to make use of these advantages with t he development of a modular-based approach to implement grayscale morphology. Since t he process of capturing the image is independent of the image processing, this modular approach has the flexibility and scalability of design to allow for non-fixed algorithms that utilize morphological operators with weighted structuring elements of variable size and shape. This is an important first step in the development of an intelligent camera