Static Image Processing Using Pynq z1

This Image Processing Project leverages the Xilinx PYNQ-Z1 board to perform real-time image processing and visualization. Utilizing Python and OpenCV, the project applies multiple image filters to a source image and outputs the results via an HDMI-connected monitor. This documentation outlines the setup, usage, code structure, and potential enhancements for the project.

Requirements:

1. Hardware: PYNQ-Z1 board, HDMI monitor, HDMI cable, 12V power supply.
2. Software: PYNQ image installed on a microSD card, Jupyter notebook environment.
3. Libraries: Python with PIL, NumPy, Matplotlib, and OpenCV installed.
4. Bitstream: Compatible base.bit file for the PYNQ-Z1 overlay.

Setup Guide:

1. Insert the microSD card with the PYNQ image into the PYNQ-Z1 board.
2. Connect the HDMI cable from the PYNQ-Z1 to the monitor and power on the board.
3. Access the Jupyter notebook server via the PYNQ-Z1’s IP address in a web browser.
4. Upload the source image SourceImage.jpg to /home/xilinx/jupyter_notebooks/Image_Processing/.
5. Ensure the base.bit file is available in the working directory.

Working:

Cell 01:

The PYNQ-Z1 Image Processing Project code operates by loading a 640x480 image, converting it from RGB to BGR for OpenCV compatibility, and displaying it in the Jupyter notebook.

CELL 02:

Initializes the PYNQ-Z1 overlay and configures the HDMI output at 640x480 with 24-bit depth.

CELL 03:

Applies filters—Grayscale, Threshold, Edge Detection, Blur, and Sharpening—using OpenCV, converts single-channel images to 3-channel BGR, and displays them in subplots.

CELL 04:

Runs an interactive loop, prompting the user to select a filter (1-5) or exit (6), then writes the chosen 3-channel BGR image to the HDMI monitor, ensuring shape and dtype compatibility with a 1-second delay for updates.

Results:

The Grayscale filtered image is then displayed on the Monitor connected via HDMI to the PYNQ board.