The design and implementation of a real-time autonomous vehicle using IoT-based image processing is presented in this research. The system incorporates a sensor camera, Raspberry Pi, and a Convolutional Neural Network (CNN) program to detect and classify various vehicles and obstacles in the surrounding environment. The captured images are transmitted to the Raspberry Pi, where a Python program executes the CNN algorithm for object identification. Upon detection, the system sends signals to the motor controller, reducing the motor speed accordingly. Additionally, voice commands are used to alert the user about the presence of objects. Ultrasonic and proximity sensors further enhance the system's capabilities by providing real-time feedback. The research results demonstrate the successful detection and classification of different vehicles, validating the effectiveness of the CNN program. The integration of motor speed control and voice commands enhances safety and user experience. The developed system holds promise for enhancing the intelligence and safety of autonomous vehicles, paving the way for further advancements in real-time object detection and response.