{"id":{"repo_id":"de-montfort","oai_identifier":"oai:dora.dmu.ac.uk:2086/25472"},"canonical_url":"https://search.dev.ndltd.org/etd/de-montfort/oai:dora.dmu.ac.uk:2086/25472","repository":{"repo_id":"de-montfort","name":"De Montfort University","base_url":"https://dora.dmu.ac.uk/server/oai/request"},"display":{"title":"Automated Visual Inspection of Lace Using Machine Vision","abstract":"The development of the modem CCD camera coupled with the vast increase in computational speed of the latest PC’s is making simple mundane visual inspection tasks increasingly cost effective. This thesis investigates the feasibility of applying machine vision to automate the inspection of lace on the knitting machine, in real time and as close to the point of knitting as possible. Developments in sensors, particularly high resolution CCD linescan cameras and in cost effective high speed processors, makes this aim increasingly feasible at an acceptable cost. Commercially available systems for automated fabric inspection, have in the past, been very expensive and has only been available for simple webs, such as paper and plain fabrics, typical systems costing well in excess of £100,000. The defects found in Raschel knitted lace were catalogued, and techniques used for the automated visual inspection of a range of web materials (particularly steel strip and plane fabric) were studied. A database of lace samples with examples of the various kinds of defect was established. Then, simple processing was tried on the lace samples, involving threshold detection, followed by non-linear binary filtering. This could however detect only hole like defects, and could not verify the fine intricate pattern of the lace. A more powerful processing methodology was developed that compares each vertical pattern repeat in the lace pattern with a perfect prototype. This comparison is a crude one and generates many false alarms which are processed out using a neural network, trained to distinguish false alarms from genuine defects. The processing scheme was ultimately implemented on a fast PC with custom built frame store, and its operation was successfully demonstrated in real time, on-line in a factory. With a 2048 element CCD camera this covered only Im of the 3.3m width of the lace, but was shown to detect all kinds of defect with less than 1 false alarm per 30m2 of lace. The frame store was custom built because no commercial frame store could be found that met the synchronisation requirements of the inspection system. This custom frame grabber is able to perform an asynchronous camera reset. The aims of the project have been met, a prototype system has been built that can successfully detect all types of defect in Raschel lace on a Im width, with a false alarm rate of less than 1 per 30m2.","abstract_html":"The development of the modem CCD camera coupled with the vast increase in computational speed of the latest PC’s is making simple mundane visual inspection tasks increasingly cost effective. This thesis investigates the feasibility of applying machine vision to automate the inspection of lace on the knitting machine, in real time and as close to the point of knitting as possible. Developments in sensors, particularly high resolution CCD linescan cameras and in cost effective high speed processors, makes this aim increasingly feasible at an acceptable cost. Commercially available systems for automated fabric inspection, have in the past, been very expensive and has only been available for simple webs, such as paper and plain fabrics, typical systems costing well in excess of £100,000. The defects found in Raschel knitted lace were catalogued, and techniques used for the automated visual inspection of a range of web materials (particularly steel strip and plane fabric) were studied. A database of lace samples with examples of the various kinds of defect was established. Then, simple processing was tried on the lace samples, involving threshold detection, followed by non-linear binary filtering. This could however detect only hole like defects, and could not verify the fine intricate pattern of the lace. A more powerful processing methodology was developed that compares each vertical pattern repeat in the lace pattern with a perfect prototype. This comparison is a crude one and generates many false alarms which are processed out using a neural network, trained to distinguish false alarms from genuine defects. The processing scheme was ultimately implemented on a fast PC with custom built frame store, and its operation was successfully demonstrated in real time, on-line in a factory. With a 2048 element CCD camera this covered only Im of the 3.3m width of the lace, but was shown to detect all kinds of defect with less than 1 false alarm per 30m2 of lace. The frame store was custom built because no commercial frame store could be found that met the synchronisation requirements of the inspection system. This custom frame grabber is able to perform an asynchronous camera reset. The aims of the project have been met, a prototype system has been built that can successfully detect all types of defect in Raschel lace on a Im width, with a false alarm rate of less than 1 per 30m2.","abstract_has_math":false,"creators":["Sanby, C."],"institution":"De Montfort University","degree_name":"PhD","degree_level":"Doctoral","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":1995,"date_issued":"1995-12","date_published":"1995-12","updated_at":"2026-07-24T06:18:37Z","subjects":[],"languages":[],"rights":[],"rights_urls":["https://dora.dmu.ac.uk/bitstreams/5721f181-f2c1-4283-b893-f098e32ab809/download"],"identifier_entries":[]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Sanby, C."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.issued","label":"Date","values":["1995-12"]},{"key":"dc:publisher.department","label":"Dc Publisher Department","values":["Faculty of Technology, Arts and Culture"]},{"key":"dc:publisher.institution","label":"Dc Publisher Institution","values":["De Montfort University"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://hdl.handle.net/2086/25472"]},{"key":"dc:type","label":"Dc Type","values":["Thesis or dissertation"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["PhD"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["https://dora.dmu.ac.uk/bitstreams/5721f181-f2c1-4283-b893-f098e32ab809/download"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://dora.dmu.ac.uk/bitstreams/c5b2ebeb-906e-4d41-9468-70ed7be6cfd7/download"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The development of the modem CCD camera coupled with the vast increase in computational speed of the latest PC’s is making simple mundane visual inspection tasks increasingly cost effective. This thesis investigates the feasibility of applying machine vision to automate the inspection of lace on the knitting machine, in real time and as close to the point of knitting as possible. Developments in sensors, particularly high resolution CCD linescan cameras and in cost effective high speed processors, makes this aim increasingly feasible at an acceptable cost. Commercially available systems for automated fabric inspection, have in the past, been very expensive and has only been available for simple webs, such as paper and plain fabrics, typical systems costing well in excess of £100,000. The defects found in Raschel knitted lace were catalogued, and techniques used for the automated visual inspection of a range of web materials (particularly steel strip and plane fabric) were studied. A database of lace samples with examples of the various kinds of defect was established. Then, simple processing was tried on the lace samples, involving threshold detection, followed by non-linear binary filtering. This could however detect only hole like defects, and could not verify the fine intricate pattern of the lace. A more powerful processing methodology was developed that compares each vertical pattern repeat in the lace pattern with a perfect prototype. This comparison is a crude one and generates many false alarms which are processed out using a neural network, trained to distinguish false alarms from genuine defects. The processing scheme was ultimately implemented on a fast PC with custom built frame store, and its operation was successfully demonstrated in real time, on-line in a factory. With a 2048 element CCD camera this covered only Im of the 3.3m width of the lace, but was shown to detect all kinds of defect with less than 1 false alarm per 30m2 of lace. The frame store was custom built because no commercial frame store could be found that met the synchronisation requirements of the inspection system. This custom frame grabber is able to perform an asynchronous camera reset. The aims of the project have been met, a prototype system has been built that can successfully detect all types of defect in Raschel lace on a Im width, with a false alarm rate of less than 1 per 30m2."]},{"key":"dc:format.checksum.md5","label":"Dc Format Checksum Md5","values":["bd41181d9a4c38b5ebacc69a027024d9","b40e124f29344173e01134a29e5ae28a","9639d6864b8674375da7dc07bfa9a215"]},{"key":"dc:title","label":"Title","values":["Automated Visual Inspection of Lace Using Machine Vision"]}]}],"canonical_facts":{"dc:creator":["Sanby, C."],"dc:date.issued":["1995-12"],"dc:description.abstract":["The development of the modem CCD camera coupled with the vast increase in computational speed of the latest PC’s is making simple mundane visual inspection tasks increasingly cost effective. This thesis investigates the feasibility of applying machine vision to automate the inspection of lace on the knitting machine, in real time and as close to the point of knitting as possible. Developments in sensors, particularly high resolution CCD linescan cameras and in cost effective high speed processors, makes this aim increasingly feasible at an acceptable cost. Commercially available systems for automated fabric inspection, have in the past, been very expensive and has only been available for simple webs, such as paper and plain fabrics, typical systems costing well in excess of £100,000. The defects found in Raschel knitted lace were catalogued, and techniques used for the automated visual inspection of a range of web materials (particularly steel strip and plane fabric) were studied. A database of lace samples with examples of the various kinds of defect was established. Then, simple processing was tried on the lace samples, involving threshold detection, followed by non-linear binary filtering. This could however detect only hole like defects, and could not verify the fine intricate pattern of the lace. A more powerful processing methodology was developed that compares each vertical pattern repeat in the lace pattern with a perfect prototype. This comparison is a crude one and generates many false alarms which are processed out using a neural network, trained to distinguish false alarms from genuine defects. The processing scheme was ultimately implemented on a fast PC with custom built frame store, and its operation was successfully demonstrated in real time, on-line in a factory. With a 2048 element CCD camera this covered only Im of the 3.3m width of the lace, but was shown to detect all kinds of defect with less than 1 false alarm per 30m2 of lace. The frame store was custom built because no commercial frame store could be found that met the synchronisation requirements of the inspection system. This custom frame grabber is able to perform an asynchronous camera reset. The aims of the project have been met, a prototype system has been built that can successfully detect all types of defect in Raschel lace on a Im width, with a false alarm rate of less than 1 per 30m2."],"dc:format.checksum.md5":["bd41181d9a4c38b5ebacc69a027024d9","b40e124f29344173e01134a29e5ae28a","9639d6864b8674375da7dc07bfa9a215"],"dc:identifier.uri":["https://dora.dmu.ac.uk/bitstreams/c5b2ebeb-906e-4d41-9468-70ed7be6cfd7/download"],"dc:publisher.department":["Faculty of Technology, Arts and Culture"],"dc:publisher.institution":["De Montfort University"],"dc:relation.isreferencedby":["https://hdl.handle.net/2086/25472"],"dc:rights":["https://dora.dmu.ac.uk/bitstreams/5721f181-f2c1-4283-b893-f098e32ab809/download"],"dc:title":["Automated Visual Inspection of Lace Using Machine Vision"],"dc:type":["Thesis or dissertation"],"dc:type.qualificationlevel":["Doctoral"],"dc:type.qualificationname":["PhD"]},"updated_at":"2026-07-24T06:18:37Z"}