An intelligent, computer vision–assisted cassava peeling system was designed, developed, and evaluated to address the limitations of manual and conventional mechanical peeling, particularly inefficiency, flesh wastage, and inability to handle irregular tuber geometry. The system integrates a microcontroller-based control unit with a vision module for contour recognition and an automated peeling mechanism. Image acquisition and processing were performed using the OpenCV library on a Raspberry Pi platform, which guided the knife’s trajectory for adaptive peeling. Performance evaluation was conducted at rotational speeds ranging from 30 to 45 rpm, feeding one tuber at a time into the peeling chamber. Results revealed that peeling efficiency increased from 82.4% to a maximum of 94.7% at 37.5 rpm, after which it slightly declined. Flesh loss decreased from 6.2% to 3.0%, and tuber breakage was minimized to 2.1% at the same optimum speed. Throughput capacity increased linearly from 15 to 22.7 tubers per hour, while power consumption rose from 65 W to 90 W across the test range. The study established 37.5 rpm as the optimal operational speed, offering a balance between efficiency, energy use, and product quality. The developed vision-guided peeler demonstrates a viable, low-energy, and high-precision approach to cassava processing, capable of enhancing productivity, reducing waste, and supporting sustainable postharvest mechanization in small- and medium-scale agro-industries.
Ogunnigbo et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: