Extraction of Agricultural Workers’ Physical Timeline Data for Acceleration and Angular Velocity through Deep Learning Based System | Chapter 7 | Science and Technology - Recent Updates and Future Prospects Vol. 6

 The present study uses timeline matrix-formed datasets with a variety of common Japanese-styled agri-tools. We base our deep learning approach on achievements in the acoustic deep learning fields, where timeline physical data is transformed into WAV formatted sound file data. Recent improvements in analyzing agri-work have considered up-to-date technologies with deep learning approaches to understand how solutions reflect the experience of traditional agri-workers.  To comprehend acceleration and angular velocity, physical timeline data can be used to derive certain physical features of workers. There is still need for improvement even if different strategies have been put into place internationally for both indoor and outdoor agricultural (agri-) working areas. We apply a deep learning-based method and qualitatively demonstrate the classification of physical timeline datasets. To create our dataset, our subjects were six experienced agri-manual workers and six completely inexperienced men. The targeted task was cultivating the semi-crunching position using a simple, Japanese-style hoe. We captured the subjects’ acceleration and angular velocity data from an integrated multi-sensor module mounted on a wood lilt 15 cm from the gripping position of the dominant hand. We used Python code and recent distributed libraries for computation. For data classification, we successively executed a Recurrent Neural Network (RNN), which we evaluated using wavelet analyses such as the Fast Fourier Transform (FFT). These methods of analyzing digital data could be of practical use for providing key suggestions to improve daily tasks. Future users could automatically or semi-automatically apply our approaches to classify a wide variety of digital matrix-formed data. In the long term, we aim to check and improve the system durability, long-term performance, and other methodological mixing patterns.

 

Author (s) Details

 

Shinji Kawakura
Kobe University, Kobe-shi, Hyogo, 657-0066, Japan.

 

Ryosuke Shibasaki
The University of Tokyo, Bunkyo-ku, Tokyo, 113-0033, Japan and Reitaku University, Kashiwa-shi, Chiba, 277-8686, Japan.

 

Please see the book here:- https://doi.org/10.9734/bpi/strufp/v6/820