Advanced Segregation of Defective Cigarettes: Integrating Image Processing and Machine Learning

This paper presents the design and development of a damaged
cigarette segregation machine, aimed at automating the process of
identifying and separating defective cigarettes from production
lines. The system integrates advanced image processing techniques
and machine learning algorithms to detect physical defects such as
broken filters, damaged wrappers, or incorrect lengths in real-time.
High-speed cameras capture images of cigarettes on the conveyor
belt, while a custom-trained neural network identifies defective
products with high accuracy. Upon detection, a pneumatic actuator
removes the damaged cigarettes from the production line. Initial
testing demonstrated an accuracy rate of over 95% in detecting
defects, significantly reducing manual inspection efforts and
improving overall production efficiency. This machine offers a
scalable, cost-effective solution for maintaining product quality
standards in the tobacco industry while minimizing human
intervention. Future work will focus on optimizing the machine's
performance and extending its application to other quality control
processes.