Visual impairment was defined as the limitation of actions and functions of the visual system. It may cause an individual's difficulty in carrying out their normal daily activities such as driving, reading, socializing, and walking. Technology innovation plays a major part in our lives today and can be utilized for many purposes and one of which is an aid for Legally Blind Persons (LBP) suffering from the defect of visual impairment. The main focus of this work is to develop an application which will help the LBP in recognizing objects, color and text using only the mobile camera. Machine learning was implemented in the study using the platform of TensorFlow. The network of algorithms or artificial neural network was also integrated into the work, thus the application has the capability of providing different interpretation of data captured by the mobile camera through deep learning. To validate the effectiveness and efficiency of work, it was evaluated by several individuals using the following criteria: Usability, Accuracy and Assistance and the result was interpreted using the Likert scale values. Overall result was 4.33 interpreted as outstanding, thus concluded that the work met the targeted individuals' expectations and requirements and that it is acceptable.