BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.osgeo.org//foss4g-asia-2023//TVJZPX
BEGIN:VTIMEZONE
TZID:KST
BEGIN:STANDARD
DTSTART:20000101T000000
RRULE:FREQ=YEARLY;BYMONTH=1
TZNAME:KST
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-foss4g-asia-2023-V8HLFY@talks.osgeo.org
DTSTART;TZID=KST:20231130T172000
DTEND;TZID=KST:20231130T174000
DESCRIPTION:Cocoa is a major global commodity\, and its production involves
  millions of mostly smallholder farmers in developing countries. Efficient
  and accurate counting of cocoa fruits during harvest is essential for eff
 ective resource management\, yield estimation\, and overall productivity i
 mprovement. Traditional manual counting methods are time-consuming\, labor
 -intensive\, and susceptible to errors\, necessitating the exploration of 
 innovative technological solutions. \n\nThis research introduces a novel a
 pproach to automatic cocoa fruit counting using a mobile application integ
 rated with computer vision techniques. The proposed system leverages four 
 services: 1) acquisition of farmers monitoring data using the mobile devic
 es\, 2) develop explainable AI model for automatic cocoa fruit recognition
 \, 3) presentation of Cocoa pod detection and counting services through a 
 web & mobile based application and 4) collect the farmers feedback on coco
 a pod count to improve the services. To do this\, Mobile app for data coll
 ection was co-designed and tested with local farmers in Cameroon. This app
  allows farmers to take geo-tagged photographs and transfer them to the se
 rver when connectivity permits. In total\, 2132 Cocoa tree photos were man
 ually annotated and different artificial intelligence (AI) models (e.g.\, 
 faster R-CNN\, darknet\, ResNet-50\, mobilenetv3) were developed and teste
 d for automatic counting of Cocoa pods. The Faster R-CNN model with a ResN
 et-50 performed better and provides accuracy above 70%. The model was depl
 oyed on a web service and provides real-time prediction through a web- & M
 obile based application \n\nThe insights gleaned from this research will s
 pawn a new generation of tools for Cocoa farmers to use. It is an efficien
 t\, non-destructive\, and low-cost method which offers useful information 
 for them to plan agricultural work and obtain economic benefits from the c
 orrect administration of resources. Nonetheless\, further research is need
 ed to address potential limitations data collection strategies (specifical
 ly variable environmental conditions\, complex Backgrounds for image colle
 ction)\, image Annotation challenges and ensure seamless integration into 
 the existing agricultural workflow.
DTSTAMP:20260413T203032Z
LOCATION:Seoul Archive
SUMMARY:Automatic Cocoa fruit count using a mobile app and computer vision 
 - Arun Kumar Pratihast\, Eliya Buyukkaya\, Jappe Franke\, Hugo de Groot
URL:https://talks.osgeo.org/foss4g-asia-2023/talk/V8HLFY/
END:VEVENT
END:VCALENDAR