{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/image-augmentation-for-object-image","title":"Image Augmentation for Object Image Classification Based On Combination of PreTrained CNN and SVM","arxiv_id":null,"date":"2018-01-01","proceeding":null,"authors":["Yoshihiro Shima"],"abstract":"Neural networks are a powerful means of classifying object images. The proposed\r\nimage category classification method for object images combines convolutional neural\r\nnetworks (CNNs) and support vector machines (SVMs). A pre-trained CNN, called Alex-Net,\r\nis used as a pattern-feature extractor. Alex-Net is pre-trained for the large-scale object-image\r\ndataset ImageNet. Instead of training, Alex-Net, pre-trained for ImageNet is used. An SVM is\r\nused as trainable classifier. The feature vectors are passed to the SVM from Alex-Net. The\r\nSTL-10 dataset are used as object images. The number of classes is ten. Training and test\r\nsamples are clearly split. STL-10 object images are trained by the SVM with data\r\naugmentation. We use the pattern transformation method with the cosine function. We also\r\napply some augmentation method such as rotation, skewing and elastic distortion. By using the\r\ncosine function, the original patterns were left-justified, right-justified, top-justified, or bottomjustified. Patterns were also center-justified and enlarged. Test error rate is decreased by 0.435\r\npercentage points from 16.055% by augmentation with cosine transformation. Error rates are\r\nincreased by other augmentation method such as rotation, skewing and elastic distortion,\r\ncompared without augmentation . Number of augmented data is 30 times that of the original\r\nSTL-10 5K training samples. Experimental test error rate for the test 8k STL-10 object images\r\nwas 15.620%, which shows that image augmentation is effective for image category\r\nclassification.","url_abs":"https://iopscience.iop.org/article/10.1088/1742-6596/1004/1/012001/pdf","url_pdf":"https://iopscience.iop.org/article/10.1088/1742-6596/1004/1/012001/pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":null,"task_name":"8k"},{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-augmentation","task_name":"Image Augmentation"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"object","task_name":"Object"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-stl-10","task":"Image Classification","dataset":"STL-10","model":"cosine function","rank_in_archive_order":51,"of":117,"metrics":{"Percentage correct":"84.38"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-stl-10","task":"Image Classification","dataset":"STL-10","model":"skewing","rank_in_archive_order":54,"of":117,"metrics":{"Percentage correct":"83.47"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-stl-10","task":"Image Classification","dataset":"STL-10","model":"elastic distortion(2)","rank_in_archive_order":55,"of":117,"metrics":{"Percentage correct":"83.45"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-stl-10","task":"Image Classification","dataset":"STL-10","model":"elastic distortion(1)","rank_in_archive_order":57,"of":117,"metrics":{"Percentage correct":"83.00"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-stl-10","task":"Image Classification","dataset":"STL-10","model":"rotation","rank_in_archive_order":60,"of":117,"metrics":{"Percentage correct":"81.45"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}