{"id":{"repo_id":"middlesex","oai_identifier":"oai:repository.mdx.ac.uk:8q395"},"canonical_url":"https://search.dev.ndltd.org/etd/middlesex/oai:repository.mdx.ac.uk:8q395","repository":{"repo_id":"middlesex","name":"Middlesex University","base_url":"https://repository.mdx.ac.uk/oai2"},"display":{"title":"Detection of human face from sketches using deep learning networks","abstract":"In the law and enforcement, sketching is a common technique to record witnesses’ version of perceived suspects. Then these sketches are employed to identify the correct persons from a photo database, which will be a time consuming task if not impossible for a human being. This work investigates the application of artificial techniques (AI), specifically, the recent state of the art deep learning techniques to identify people based on their sketches for law and enforcement. The dataset is collected from Chinese database (CUHK student dataset) that includes both photos and their corresponding sketches. The sketches are drawn by artists. The deep learning network is Siamese network, which uses Python programming language. Since the data base contains only less than 200 photos, data enlargement is also conducted to generate sketch images from face images, which generates another 500 pairs of images. As a result, the Siamese network achieved 96% accuracy when retrieving photos from sketch images. Future work include further enlarging the database and evaluating other deep learning architectures.","abstract_html":"In the law and enforcement, sketching is a common technique to record witnesses’ version of perceived suspects. Then these sketches are employed to identify the correct persons from a photo database, which will be a time consuming task if not impossible for a human being. This work investigates the application of artificial techniques (AI), specifically, the recent state of the art deep learning techniques to identify people based on their sketches for law and enforcement. The dataset is collected from Chinese database (CUHK student dataset) that includes both photos and their corresponding sketches. The sketches are drawn by artists. The deep learning network is Siamese network, which uses Python programming language. Since the data base contains only less than 200 photos, data enlargement is also conducted to generate sketch images from face images, which generates another 500 pairs of images. As a result, the Siamese network achieved 96% accuracy when retrieving photos from sketch images. Future work include further enlarging the database and evaluating other deep learning architectures.","abstract_has_math":false,"creators":["Abbey, M."],"institution":"Middlesex University","degree_name":null,"degree_level":"Masters thesis","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022","date_published":"2022","updated_at":"2026-07-24T03:03:59Z","subjects":[],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:repository.mdx.ac.uk:8q395"],"render_values":[{"text":"oai:repository.mdx.ac.uk:8q395","href":null,"code":true}]}]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Abbey, M."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2022"]},{"key":"dc:date.issued","label":"Date","values":["2022"]},{"key":"dc:publisher.department","label":"Dc Publisher Department","values":["Computer Science"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["Middlesex University"]},{"key":"dc:relation","label":"Dc Relation","values":["https://repository.mdx.ac.uk/item/8q395"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://repository.mdx.ac.uk/item/8q395"]},{"key":"dc:type","label":"Dc Type","values":["Thesis or dissertation"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Masters thesis"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:repository.mdx.ac.uk:8q395"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://repository.mdx.ac.uk/download/adceaffda4ada4b289af1f07285a34a9249cd6df89e3c75a798bee0aa80432d4/5488650/MAbbey%20thesis.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["In the law and enforcement, sketching is a common technique to record witnesses’ version of perceived suspects. Then these sketches are employed to identify the correct persons from a photo database, which will be a time consuming task if not impossible for a human being. This work investigates the application of artificial techniques (AI), specifically, the recent state of the art deep learning techniques to identify people based on their sketches for law and enforcement. The dataset is collected from Chinese database (CUHK student dataset) that includes both photos and their corresponding sketches. The sketches are drawn by artists. The deep learning network is Siamese network, which uses Python programming language. Since the data base contains only less than 200 photos, data enlargement is also conducted to generate sketch images from face images, which generates another 500 pairs of images. As a result, the Siamese network achieved 96% accuracy when retrieving photos from sketch images. Future work include further enlarging the database and evaluating other deep learning architectures."]},{"key":"dc:description.abstract","label":"Abstract","values":["In the law and enforcement, sketching is a common technique to record witnesses’ version of perceived suspects. Then these sketches are employed to identify the correct persons from a photo database, which will be a time consuming task if not impossible for a human being. This work investigates the application of artificial techniques (AI), specifically, the recent state of the art deep learning techniques to identify people based on their sketches for law and enforcement. The dataset is collected from Chinese database (CUHK student dataset) that includes both photos and their corresponding sketches. The sketches are drawn by artists. The deep learning network is Siamese network, which uses Python programming language. Since the data base contains only less than 200 photos, data enlargement is also conducted to generate sketch images from face images, which generates another 500 pairs of images. As a result, the Siamese network achieved 96% accuracy when retrieving photos from sketch images. Future work include further enlarging the database and evaluating other deep learning architectures."]},{"key":"dc:title","label":"Title","values":["Detection of human face from sketches using deep learning networks"]}]}],"canonical_facts":{"dc:creator":["Abbey, M."],"dc:date":["2022"],"dc:date.issued":["2022"],"dc:description":["In the law and enforcement, sketching is a common technique to record witnesses’ version of perceived suspects. Then these sketches are employed to identify the correct persons from a photo database, which will be a time consuming task if not impossible for a human being. This work investigates the application of artificial techniques (AI), specifically, the recent state of the art deep learning techniques to identify people based on their sketches for law and enforcement. The dataset is collected from Chinese database (CUHK student dataset) that includes both photos and their corresponding sketches. The sketches are drawn by artists. The deep learning network is Siamese network, which uses Python programming language. Since the data base contains only less than 200 photos, data enlargement is also conducted to generate sketch images from face images, which generates another 500 pairs of images. As a result, the Siamese network achieved 96% accuracy when retrieving photos from sketch images. Future work include further enlarging the database and evaluating other deep learning architectures."],"dc:description.abstract":["In the law and enforcement, sketching is a common technique to record witnesses’ version of perceived suspects. Then these sketches are employed to identify the correct persons from a photo database, which will be a time consuming task if not impossible for a human being. This work investigates the application of artificial techniques (AI), specifically, the recent state of the art deep learning techniques to identify people based on their sketches for law and enforcement. The dataset is collected from Chinese database (CUHK student dataset) that includes both photos and their corresponding sketches. The sketches are drawn by artists. The deep learning network is Siamese network, which uses Python programming language. Since the data base contains only less than 200 photos, data enlargement is also conducted to generate sketch images from face images, which generates another 500 pairs of images. As a result, the Siamese network achieved 96% accuracy when retrieving photos from sketch images. Future work include further enlarging the database and evaluating other deep learning architectures."],"dc:identifier":["oai:repository.mdx.ac.uk:8q395"],"dc:identifier.uri":["https://repository.mdx.ac.uk/download/adceaffda4ada4b289af1f07285a34a9249cd6df89e3c75a798bee0aa80432d4/5488650/MAbbey%20thesis.pdf"],"dc:publisher.department":["Computer Science"],"dc:publisher.institution":["Middlesex University"],"dc:relation":["https://repository.mdx.ac.uk/item/8q395"],"dc:relation.isreferencedby":["https://repository.mdx.ac.uk/item/8q395"],"dc:title":["Detection of human face from sketches using deep learning networks"],"dc:type":["Thesis or dissertation"],"dc:type.qualificationlevel":["Masters thesis"]},"updated_at":"2026-07-24T03:03:59Z"}