Visual impairment affects over 2.2 billion people globally [1], creating significant challenges in the life of visually impaired people. Object detection and recognition are crucial in developing a model to aid in independent navigation of visually impaired individuals. Object detection refers to the detection of obstacles in the surroundings of visually impaired individuals, and recognition includes the identification of the type of object. Modern technologies like Computer Vision and Deep Learning can enhance the model’s efficacy. In object detection and recognition, the YOLO(You Only Look Once) model holds supremacy by detecting the objects in a single convolutional network with high accuracy. It also leads to ease of deployment and speed of detection. This research aims to determine the best Yolov8 learning rate by fine-tuning Yolov8 with five evolutionary algorithms and by making a comparative analysis. Evolutionary algorithms fine-tune the model and make the model befitting for object detection, object recognition in a better approach. The model achieves an accuracy of 70.1% in object detection and recognition after fine-tuning the Yolov8 model with Genetic Algorithm marking a significant advancement in assistive technology and ensures an invaluable contribution to the realm of Computer Vision.