This study investigates the performance of the advance waste sorting system employing object detection technique to classify plastic bottles as either recyclable or nonrecyclable. The system integrated with Python-based object detection models, which was trained to distinguish between biodegradable and nonbiodegradable plastic bottles. The trained model was optimized and converted into a format compatible with the ESP32 microcontroller, facilitating real-time waste classification and sorting. The series of experiments were conducted including five sets of tests with 30 trails each to refine the detection accuracy. Result showed that with accurate camera placements, system detecting recyclable bottles based on logo identification efficiency reached 88.5%. Furthermore, the integration of object detection with microcontroller communication via the Arduino IDE enhanced recycling processes and managing plastic waste effectively.