Toward semantic action recognition for avocado harvesting process based on single shot multibox detector

Juan Pablo Vasconez, Jaime Salvo, Fernando Auat

Research output: Chapter in Book/Report/Conference proceedingConference contribution

10 Citations (Scopus)

Abstract

To date, human action recognition is still a challenging topic and has been addressed from many perspectives. Detection of human actions can be useful to obtain relevant information to improve complex processes, which is the case of agricultural applications. In this work, the detection of objects that can provide information for human action recognition based on semantic representations is studied. For this purpose, a convolutional neuronal network based on Single Shot MultiBox Detector meta-architecture and MobileNet feature extractor was implemented, which has been trained to detect nine classes of objects during the process of collecting avocados in a Chilean farm. We have found that such detected objects are related to seven possible actions that can be detected during avocado harvesting process. Such information could allow to directly detect certain actions in still images, or improve conventional action detection methods during the harvesting process. The results show that is possible to detect human actions during the process, obtaining action recognition performances from 41% to 80% depending on the task. This approach can help to obtain information about how to improve harvesting process and reduce human workload in near future, which may be an important contribution for the search of sustainable agricultural practices.

Original languageEnglish
Title of host publication2018 IEEE International Conference on Automation/23rd Congress of the Chilean Association of Automatic Control
PublisherIEEE
ISBN (Electronic)9781538655863
DOIs
Publication statusPublished - 13 Jan 2019
Event2018 IEEE International Conference on Automation/23rd Congress of the Chilean Association of Automatic Control - Greater Concepcion, Chile
Duration: 17 Oct 201819 Oct 2018

Conference

Conference2018 IEEE International Conference on Automation/23rd Congress of the Chilean Association of Automatic Control
Abbreviated titleICA-ACCA 2018
Country/TerritoryChile
CityGreater Concepcion
Period17/10/1819/10/18

Keywords

  • agriculture
  • avocado harvesting
  • Semantic action recognition
  • Single Shot MultiBox Detector

ASJC Scopus subject areas

  • Artificial Intelligence
  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering
  • Control and Optimization
  • Instrumentation

Fingerprint

Dive into the research topics of 'Toward semantic action recognition for avocado harvesting process based on single shot multibox detector'. Together they form a unique fingerprint.

Cite this