A framework for performance evaluation of plucking activity in tea

Cultivation of tea is characterized by a number of activities spread over many months in which plucking is a major activity. In north-eastern and southern parts of India where tea is a major crop, plucking involves picking of two leaves and a bud with bare hands which after careful processing is converted into the tea beverage. However, no precise ways to characterize and quantify such activities have been developed. We propose a novel framework to characterize a tea-worker's plucking performance with the help of wrist-borne wearable systems. With a limited observation period in an uncontrolled environment for a tea-worker, our framework obtains (a) with a high accuracy the pluck-count which is directly associated with leaf-count and (b) other useful measures associated with related sub-activities. We believe this is a harbinger of change which will go a long way in improving the overall efficiency of the tea industry.