Published Paper Details:

This project aims to design a system utilizing FPGA for the classification of glaucoma disease using a CNNDENSENET121 based deep learning model. The primary goal is to optimize the design for the Airtex V7 FPGA architecture, employing a graphical user interface (GUI) for user interaction. The project begins by processing a glaucoma image dataset, where images are pre-processed through resizing, noise removal, normalization, and binary pattern extraction. Threshold segmentation is then applied to detect the affected regions in the images. Feature extraction is carried out using a CNNDENSENET121 model to identify critical patterns indicative of glaucoma. The dataset is split into training and testing subsets, with 80% allocated for model evaluation and 20% for prediction. For classification, the system incorporates the CNNDENSENET121 algorithm, which includes convolutional layers, batch normalization, activation functions (ReLU), pooling layers, and fully connected layers. The FPGA design is optimized for performance, with the final output prediction generated on the FPGA hardware. The system's performance is assessed using various metrics such as area, power, delay, and accuracy, along with RTL simulation and synthesis reports. Additional metrics like PSNR, SSIM, MSE, MAE, precision, recall, ROC, confusion matrix, and error rate are also evaluated to ensure robust classification performance. The proposed system effectively demonstrates the integration of machine learning algorithms with FPGA hardware for efficient glaucoma detection.