1 Introduction1.1 Problem Description Nowadays, disadvantaged people such as visually impaired people are found all over the world. According to an estimate by the World Health Organization (WHO) in 2013, 285 million people worldwide are visually impaired. Additionally, 39 million people in the population are blind. These people need help in their daily lives, such as locating and grabbing objects (e.g. a cup, a key, etc.) that may have been dropped or misplaced. In fact, they have less information about the environment. Therefore, the development of assistive technologies and wearable devices can help them increase their independence and reduce inconvenience. The ultimate goal is to improve their lives with today's advanced technology.1.2 MotivationVisually impaired people encounter inconveniences when interacting with their environment. The hardest part is finding a specific item. This proposed system is designed in such a way as to help visually impaired people to increase the ease of manual navigation to locate and grasp their necessities in an environment. Computer vision and image processing play an important role in this system. Images contain a lot of information that can be used to help them fully understand the environment around them. Additionally, each object has its own local features such as regions, blobs, and points. By using feature points in images, it is easier to recognize objects that can help the visually impaired. Therefore, by using this method, visually impaired people can reduce their disadvantages and improve their lives.1.3 Proposed ApproachFigure 1 Feature point search process. Typically, object detection is simplified such that the system focuses on a predetermined invariant SURF. ..... middle of paper ......ems into daily life with a portable webcam to reduce inconvenience. Additionally, it can guide the user's hand to locate and grasp their essential objects in an environment. The main objective of this system is to use feature points for object recognition. The Speed-Up Robust Feature (SURF) is rotation and scale invariant. It can handle image translation, scaling, changing viewpoints, and rotating between objects in a cluttered background. Thus, it is used as a feature detector and feature descriptor for this system.6.1 Future WorkFuture work will focus on improving the object recognition system so that it can better identify and detect objects under different conditions. Additionally, a human-machine interface (HCI) will be included in the system for auditory display and image capture of object recognition on smartphone and computer..