virtual impaired system of personal dataset

The personal data is used to create a dataset of images containing its items

The objective of personal data for blind classification systems is to develop a computer vision system that is capable of recognizing and classifying personal data that blind individuals commonly use and to help them locate their personal belongings more easily and independently by enabling a deep learning model to learn the visual features and patterns of each item, and he has enough pictures to capture the variability of each object, such as different angles, lighting conditions, and backgrounds, this can help to blind individuals to locate their personal belongings easily and independently.

The dataset focuses on personal items used by blind individuals, it was collected by capturing images of common items used by blind, It is 10 classes My bag, My cup, My wallet, My toothbrush, My sunglasses, and the corresponding items for each class. Images were captured under various lighting conditions and angles for variability and are labeled accordingly.

The dataset is divided into training, testing, and validation sets, with images resized to 1000 by 1000 pixels to reduce computational complexity and speed up training.

.I use Google Colab to train model and use tensorflow, Keras, and sklearn library

The model architecture includes an Input layer, multiple convolutional layers for feature extraction, pooling layers to reduce dimensionality, fully connected layers for processing, and an Output layer for classification.