{"id":{"repo_id":"iastate","oai_identifier":"oai:dr.lib.iastate.edu:20.500.12876/azJ4WgWv"},"canonical_url":"https://search.dev.ndltd.org/etd/iastate/oai:dr.lib.iastate.edu:20.500.12876/azJ4WgWv","repository":{"repo_id":"iastate","name":"Iowa State University","base_url":"https://dr.lib.iastate.edu/server/oai/request"},"display":{"title":"Development of image processing algorithms for automatic segmentation and selection of swine postural behaviors used in an interactive environmental control","abstract":"Traditionally, environmental control in swine production facilities is air temperature-based. Although it can protect the well-being of pigs to a certain degree and is fairly easy to operate, it may fail to ensure the comfort environments even though seemingly desired temperatures can be maintained precisely. The reason lies in its inherent incapability to integrate all the effects of the factors that influence the thermal needs of the animals. A novel swine postural behavior-based environmental control scheme was proposed to reflect and respond to this thermal need. Previous research has shown that this method possesses the potential to achieve a better interactive environmental control. The challenges from image segmentation and image selection of swine postural behavior for this control scheme require the study reported in this thesis. Image segmentation algorithm based on multithresholding was employed with the assistance of background reference technique. Background reference eliminates the effects of the entire background which if not handled properly may ruin the efficiency of multithresholding. This method was capable of producing satisfactory segmentation result when both black and white pigs with different intensity distributions were present at the same time, which always challenges those methods developed for general purpose. Selection of wall/floor color facilitated the proposed segmentation method by augmenting the intensity difference between background pixels and pixels belonging to pigs. Guidelines for color selection were obtained by experiment and proved practical. Reference image updating algorithm based on statistical hypothesis testing was developed to trace the changes in the background so that the influences of changing illumination, color fading and shadowing would have little effect on the performance of the subsequent image segmentation. Only the swine postural behavior images are the desired images which effectively reveal the comfortable state of animals. To select this type of image before applying image segmentation algorithm, image motion detection algorithm was developed to eliminate the image sequence of moving objects. This method combined the advantages of likelihood ratio method and shading model method and showed stable performance under noisy and illumination changing conditions. Further research to improve the performance of the image processing algorithms is desired.","abstract_html":"Traditionally, environmental control in swine production facilities is air temperature-based. Although it can protect the well-being of pigs to a certain degree and is fairly easy to operate, it may fail to ensure the comfort environments even though seemingly desired temperatures can be maintained precisely. The reason lies in its inherent incapability to integrate all the effects of the factors that influence the thermal needs of the animals. A novel swine postural behavior-based environmental control scheme was proposed to reflect and respond to this thermal need. Previous research has shown that this method possesses the potential to achieve a better interactive environmental control. The challenges from image segmentation and image selection of swine postural behavior for this control scheme require the study reported in this thesis. Image segmentation algorithm based on multithresholding was employed with the assistance of background reference technique. Background reference eliminates the effects of the entire background which if not handled properly may ruin the efficiency of multithresholding. This method was capable of producing satisfactory segmentation result when both black and white pigs with different intensity distributions were present at the same time, which always challenges those methods developed for general purpose. Selection of wall/floor color facilitated the proposed segmentation method by augmenting the intensity difference between background pixels and pixels belonging to pigs. Guidelines for color selection were obtained by experiment and proved practical. Reference image updating algorithm based on statistical hypothesis testing was developed to trace the changes in the background so that the influences of changing illumination, color fading and shadowing would have little effect on the performance of the subsequent image segmentation. Only the swine postural behavior images are the desired images which effectively reveal the comfortable state of animals. To select this type of image before applying image segmentation algorithm, image motion detection algorithm was developed to eliminate the image sequence of moving objects. This method combined the advantages of likelihood ratio method and shading model method and showed stable performance under noisy and illumination changing conditions. Further research to improve the performance of the image processing algorithms is desired.","abstract_has_math":false,"creators":["Hu, Jianing"],"institution":null,"degree_name":"Master of Science","degree_level":"Masters","degree_discipline":"Agricultural and Biosystems Engineering","degree_department":null,"school":null,"contributors":[],"advisors":["Xin, Hongwei"],"committee_chairs":[],"committee_members":[],"year":1998,"date_issued":"1998","date_published":"1998","updated_at":"2026-07-24T02:38:57Z","subjects":[],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://dr.lib.iastate.edu/handle/20.500.12876/azJ4WgWv","outbound_label":"Repository record","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Xin, Hongwei"]},{"key":"dc:creator","label":"Author","values":["Hu, Jianing"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2024-12-20T17:11:01Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2024-12-20T17:11:01Z"]},{"key":"dc:date.issued","label":"Date","values":["1998"]},{"key":"dc:type","label":"Dc Type","values":["thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Agricultural and Biosystems Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://dr.lib.iastate.edu/handle/20.500.12876/azJ4WgWv"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Traditionally, environmental control in swine production facilities is air temperature-based. Although it can protect the well-being of pigs to a certain degree and is fairly easy to operate, it may fail to ensure the comfort environments even though seemingly desired temperatures can be maintained precisely. The reason lies in its inherent incapability to integrate all the effects of the factors that influence the thermal needs of the animals. A novel swine postural behavior-based environmental control scheme was proposed to reflect and respond to this thermal need. Previous research has shown that this method possesses the potential to achieve a better interactive environmental control. The challenges from image segmentation and image selection of swine postural behavior for this control scheme require the study reported in this thesis. Image segmentation algorithm based on multithresholding was employed with the assistance of background reference technique. Background reference eliminates the effects of the entire background which if not handled properly may ruin the efficiency of multithresholding. This method was capable of producing satisfactory segmentation result when both black and white pigs with different intensity distributions were present at the same time, which always challenges those methods developed for general purpose. Selection of wall/floor color facilitated the proposed segmentation method by augmenting the intensity difference between background pixels and pixels belonging to pigs. Guidelines for color selection were obtained by experiment and proved practical. Reference image updating algorithm based on statistical hypothesis testing was developed to trace the changes in the background so that the influences of changing illumination, color fading and shadowing would have little effect on the performance of the subsequent image segmentation. Only the swine postural behavior images are the desired images which effectively reveal the comfortable state of animals. To select this type of image before applying image segmentation algorithm, image motion detection algorithm was developed to eliminate the image sequence of moving objects. This method combined the advantages of likelihood ratio method and shading model method and showed stable performance under noisy and illumination changing conditions. Further research to improve the performance of the image processing algorithms is desired."]},{"key":"dc:title","label":"Title","values":["Development of image processing algorithms for automatic segmentation and selection of swine postural behaviors used in an interactive environmental control"]}]}],"canonical_facts":{"dc:contributor.advisor":["Xin, Hongwei"],"dc:creator":["Hu, Jianing"],"dc:date.accessioned":["2024-12-20T17:11:01Z"],"dc:date.available":["2024-12-20T17:11:01Z"],"dc:date.issued":["1998"],"dc:description.abstract":["Traditionally, environmental control in swine production facilities is air temperature-based. Although it can protect the well-being of pigs to a certain degree and is fairly easy to operate, it may fail to ensure the comfort environments even though seemingly desired temperatures can be maintained precisely. The reason lies in its inherent incapability to integrate all the effects of the factors that influence the thermal needs of the animals. A novel swine postural behavior-based environmental control scheme was proposed to reflect and respond to this thermal need. Previous research has shown that this method possesses the potential to achieve a better interactive environmental control. The challenges from image segmentation and image selection of swine postural behavior for this control scheme require the study reported in this thesis. Image segmentation algorithm based on multithresholding was employed with the assistance of background reference technique. Background reference eliminates the effects of the entire background which if not handled properly may ruin the efficiency of multithresholding. This method was capable of producing satisfactory segmentation result when both black and white pigs with different intensity distributions were present at the same time, which always challenges those methods developed for general purpose. Selection of wall/floor color facilitated the proposed segmentation method by augmenting the intensity difference between background pixels and pixels belonging to pigs. Guidelines for color selection were obtained by experiment and proved practical. Reference image updating algorithm based on statistical hypothesis testing was developed to trace the changes in the background so that the influences of changing illumination, color fading and shadowing would have little effect on the performance of the subsequent image segmentation. Only the swine postural behavior images are the desired images which effectively reveal the comfortable state of animals. To select this type of image before applying image segmentation algorithm, image motion detection algorithm was developed to eliminate the image sequence of moving objects. This method combined the advantages of likelihood ratio method and shading model method and showed stable performance under noisy and illumination changing conditions. Further research to improve the performance of the image processing algorithms is desired."],"dc:identifier.uri":["https://dr.lib.iastate.edu/handle/20.500.12876/azJ4WgWv"],"dc:language.iso":["en"],"dc:title":["Development of image processing algorithms for automatic segmentation and selection of swine postural behaviors used in an interactive environmental control"],"dc:type":["thesis"],"thesis:degree_discipline":["Agricultural and Biosystems Engineering"],"thesis:degree_level":["Masters"],"thesis:degree_name":["Master of Science"]},"updated_at":"2026-07-24T02:38:57Z"}