{"corpus_id":6796548,"paper_sha":"ef0816ce5ec99ab4e2ad39df0017d463f9d4abcd","doi":"10.1109/ICRA.2017.7989612","arxiv_id":"1702.01247","pmid":null,"pmcid":null,"mag_id":2737162192,"dblp_id":"conf/icra/HallDKM17","acl_id":null,"title":"Towards unsupervised weed scouting for agricultural robotics","year":2017,"publication_date":"2017-02-04","venue":"IEEE International Conference on Robotics and Automation","journal":{"name":"2017 IEEE International Conference on Robotics and Automation (ICRA)","pages":"5223-5230","volume":null},"journal_issn":null,"journal_title":null,"publication_types":["JournalArticle","Conference"],"pubmed_pub_types":null,"s2_fields_of_study":["Agricultural and Food Sciences","Computer Science","Engineering"],"reference_count":37,"citation_count":28,"influential_citation_count":0,"is_open_access":true,"arxiv_categories":["cs.CV"],"arxiv_license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","arxiv_journal_ref":null,"mesh_headings":null,"chemicals":null,"comments_corrections":null,"source_flags":1,"s2_open_access_pdf_url":"http://arxiv.org/pdf/1702.01247","s2_open_access_landing_url":"https://www.semanticscholar.org/paper/ef0816ce5ec99ab4e2ad39df0017d463f9d4abcd","s2_open_access_license":null,"s2_open_access_status":"GREEN","pmc_open_access_pdf_url":null,"pmc_open_access_landing_url":null,"pmc_open_access_license":null,"pmc_open_access_status":null,"unpaywall_open_access_pdf_url":null,"unpaywall_open_access_landing_url":null,"unpaywall_open_access_license":null,"unpaywall_open_access_status":null,"abstract":"Weed scouting is an important part of modern integrated weed management but can be time consuming and sparse when performed manually. Automated weed scouting and weed destruction has typically been performed using classification systems able to classify a set group of species known a priori. This greatly limits deployability as classification systems must be retrained for any field with a different set of weed species present within them. In order to overcome this limitation, this paper works towards developing a clustering approach to weed scouting which can be utilized in any field without the need for prior species knowledge. We demonstrate our system using challenging data collected in the field from an agricultural robotics platform. We show that considerable improvements can be made by (i) learning low-dimensional (bottleneck) features using a deep convolutional neural network to represent plants in general and (ii) tying views of the same area (plant) together. Deploying this algorithm on in-field data collected by AgBotII, we are able to successfully cluster cotton plants from grasses without prior knowledge or training for the specific plants in the field.","claims":[{"public_id":"cl_df886a44100b0181d26b3c673b3425ac","status":"active","text":"A clustering algorithm deployed on AgBotII in-field data successfully clusters cotton plants from grasses without prior species knowledge or field-specific training.","confidence":0.92,"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/claims/cl_df886a44100b0181d26b3c673b3425ac"},{"public_id":"cl_5fb36c46a3fed282c816476b16b89e26","status":"active","text":"Learning low-dimensional bottleneck features using a deep convolutional neural network makes considerable improvements to unsupervised weed clustering performance.","confidence":0.87,"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/claims/cl_5fb36c46a3fed282c816476b16b89e26"},{"public_id":"cl_f4bc8ec93dfca9f940077d732c2c2cac","status":"active","text":"Tying views of the same plant together makes considerable improvements to the weed clustering approach.","confidence":0.87,"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/claims/cl_f4bc8ec93dfca9f940077d732c2c2cac"}],"concepts":[{"public_id":"co_2bd05ac956bdf2c18c4f8c588f723502","status":"active","name":"grasses","description":"The weed type clustered separately from cotton plants in the in-field evaluation.","types":["organism"],"aliases":[],"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_2bd05ac956bdf2c18c4f8c588f723502"},{"public_id":"co_5beec4436b35977a20064c0f47c60c29","status":"active","name":"weed scouting","description":"The practice of identifying and mapping weed species in agricultural fields to support weed management decisions.","types":["agricultural task"],"aliases":[],"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_5beec4436b35977a20064c0f47c60c29"},{"public_id":"co_9888cab77f9f8f4c4632e8707d9ca3d1","status":"active","name":"bottleneck features","description":"Low-dimensional compressed representations of plant images learned by the deep CNN for downstream clustering.","types":["representation"],"aliases":["low-dimensional features"],"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_9888cab77f9f8f4c4632e8707d9ca3d1"},{"public_id":"co_a58cff73c19afa3f27aa62e84a6cf0d2","status":"active","name":"view tying","description":"A technique that links multiple observations of the same plant area together to improve clustering coherence.","types":["method"],"aliases":[],"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_a58cff73c19afa3f27aa62e84a6cf0d2"},{"public_id":"co_b53fa22694a0d2a11053783a79f90e79","status":"active","name":"AgBotII","description":"An agricultural robotics platform used in this study to collect in-field plant image data for weed scouting.","types":["system"],"aliases":[],"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_b53fa22694a0d2a11053783a79f90e79"},{"public_id":"co_b87f3f2024247cd0aac47244253f211c","status":"active","name":"unsupervised weed clustering","description":"A machine learning approach that groups plant images into species-based clusters without labeled training data or prior species knowledge.","types":["method"],"aliases":[],"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_b87f3f2024247cd0aac47244253f211c"},{"public_id":"co_b8fe94f4cc0ecbf536f4d660564ef30d","status":"active","name":"cotton plants","description":"The crop species successfully distinguished from grasses in the unsupervised clustering experiments.","types":["organism"],"aliases":["cotton"],"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_b8fe94f4cc0ecbf536f4d660564ef30d"},{"public_id":"co_d9f72f07f0985f2ab59141c25783a98c","status":"active","name":"deep convolutional neural network","description":"A neural network architecture used here to learn compact low-dimensional feature representations of plant images.","types":["model"],"aliases":["deep CNN"],"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_d9f72f07f0985f2ab59141c25783a98c"},{"public_id":"co_eb6c5fad8ad35722b61f139ac089a39a","status":"active","name":"integrated weed management","description":"An agricultural strategy that combines multiple methods to control weed populations in crop fields.","types":["agricultural practice"],"aliases":["IWM"],"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_eb6c5fad8ad35722b61f139ac089a39a"},{"public_id":"co_f2c12b0a642b36592ab8de1db370877a","status":"active","name":"species classification","description":"A supervised weed identification approach that requires prior knowledge of the specific plant species present and must be retrained for new fields.","types":["method"],"aliases":[],"contributors":[{"id":17,"public_id":"322360f1c1","public_label":"Killer Whale (322360f1c1)","roles":["extraction"],"url":"https://sah.borca.ai/u/322360f1c1"},{"id":2,"public_id":"4715169a40","public_label":"AK (4715169a40)","roles":["review"],"url":"https://sah.borca.ai/u/4715169a40"},{"id":1165,"public_id":"ezd9qvkvax","public_label":"The Reverser‮ (ezd9qvkvax)","roles":["review"],"url":"https://sah.borca.ai/u/ezd9qvkvax"}],"url":"https://sah.borca.ai/concepts/co_f2c12b0a642b36592ab8de1db370877a"}],"external_ids":{"DOI":"10.1109/ICRA.2017.7989612","ArXiv":"1702.01247","PubMed":null,"PubMedCentral":null,"MAG":2737162192,"DBLP":"conf/icra/HallDKM17","ACL":null},"open_access":{"is_open_access":true,"pdf_url":"http://arxiv.org/pdf/1702.01247","landing_url":"https://www.semanticscholar.org/paper/ef0816ce5ec99ab4e2ad39df0017d463f9d4abcd","source":"semantic_scholar","pdf_url_source":"semantic_scholar_open_access_pdf","license":null,"status":"GREEN","reason":null},"reference_availability":{"status":"available","references_indexed":true,"full_text_available":true,"full_text_source":"arxiv","count_basis":"semantic_scholar_metadata","extraction_status":"not_applicable","reason":null},"source":{"provider":"episteme2","base_corpus":"semantic_scholar_dump","freshness_mode":"unknown","basis":["semantic_scholar_metadata","postgres_metadata"],"limits":["paper metadata is based on indexed upstream scholarly datasets","claims and concepts are available only for extracted papers","absence of claims or concepts means no extracted graph data is available in this response"],"status":"available","degraded":false,"degraded_reasons":[],"diagnostics":{"status":"available","degraded":false,"degraded_reasons":[],"metadata_status":"available","graph_status":"available","abstract_status":"available"},"source_flags":1},"paper_id":630926,"paper_uid":"6d41e4d1-260a-4b2f-af98-38935c604ff1","canonical_identity":{"paper_id":630926,"paper_uid":"6d41e4d1-260a-4b2f-af98-38935c604ff1","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/6796548"}