Cost Effective Inspection System For Automated Large Scale Cocoon Quality Assessment
Project Description :

Silk, as a textile material, has 5000 years of history and recently found newer applications in the field of biomedical engineering, battery technology, bio-sensing, humidity sensing, steam engines, waste heat management, cell culture and tissue engineering. due to the lower quality of silk produced, indian silk is less exported as compared to chinese silk. to achieve global competitiveness, india needs to bring in quality consciousness among the cocoon farmers as well as reelers. the cocoon quality directly controls the silk quality and it is currently assessed by skilled laborers manually for a few random lots, due to pausity of time between arrival and auction. this method is prone to human errors, skilled laborer dependent and time-consuming. these errors result in the loss to both the farmers and reelers in terms of price not commensurated with the quality. this problem can be addressed by developing an automated cocoon quality assessment system for grading and pricing of cocoons based on the quality. this project demonstrates an innovative automated quality assessment system for cocoons, that offers significant advantages over the manual method in terms of accuracy, userfriendliness, speed and cost effectiveness. the developed system consists of tandomly operated twin independent sections, namely image processing and sound processing. image processing section consists of the conditioned illumination unit, image acquisition and processing unit realized with smart camera which acquires the images of cocoons in a given lot and assessing the cocoon quality by running the custom-made cross-platform image processing algorithms on acquired images. by quantitative measurement of size, shape and stain color of the cocoons this system can automatically classify each cocoon in all the lots into not only good cocoons and defective ones but also defective ones into double, externally stained, uzi pierced and urinated ones. this system highlights the defective cocoons on camera screen along with statistical information such as counts of all cocoons, good ones, defective ones in each category and defect percentage at a processing speed of 96 cocoons per second with 100 % accuracy as compared to 1 cocoon per second with lesser accuracy in the manual method. in addition, the system is programmed to alert the user when the defect percentage exceeds a particular threshold value, therefore it is helpful in deciding the grade of cocoons based on the percent of defective cocoons. however, the stains on the unexposed area cannot be detected by this method and this will adversely affect the system accuracy. to overcome this, cocoons are rolled over a slope one by one at the speed of four rotations per second while the system captures and processes the video of the whole surface of the cocoons. the defective cocoons such as immature cocoons, mute, melt and calcified cocoons which are not detected by image processing section are detected by the sound processing section of the same system by analyzing the frequency and intensity of sound produced by cocoons on vibrating. sound processing section consists of vibration unit realized with servo motor and arduino board, sound capture unit and sound processing unit realized with 2 microphones and general purpose processor respectively. by employing these two sections most of the critically important defective cocoon types are identified, quantified and could be easily sorted out mechanically. this automated cocoon quality analyzing system would have a greater impact on global silk industry, especially in india, in terms of introduction of scientific cocoon quality assessment and pricing without subjectivity. this system is amenable to be integrated with a cocoon sorting machine to sort out all defective ones so that only the good cocoons will be transacted fairly at a premium prize.

Other Photos :

No Updates

Project Details :
  • Date : Nov 25,2016
  • Innovator : Prasobhkumar P. P.
  • Guide Name : Dr. Sai Siva Gorthi And Prof. C. R. Francis
  • University : Indian Institute of Science
  • Submission Year : 2017
  • Category : Agricultural Engineering
Share Project :