[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-188189-en":3,"doc-seo-188189-105":30,"detail-sidebar-cat-1-en-105":95},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":11,"category_id":12,"category_name":13,"doc_title":14,"doc_description":15,"doc_content":16,"file_id":17,"file_url":18,"file_type":19,"file_size":20,"view_count":11,"is_deleted":4,"is_public":11,"is_downloadable":11,"audit_status":11,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":15,"update_tm":28,"read_time":29},188189,1374391974468,"Eden","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",1,158,"General","Dang FEHash-Full Entropy Hash for Face Template Protection","This paper introduces a novel approach called Full Entropy Hash (FEHash) for robust and secure face template protection. The proposed method leverages random feature sampling and a defined hashing function to generate a binary string representation of facial features, which is then cryptographically hashed for secure storage. The FEHash method is evaluated on standard face recognition databases such as PIE, FEI, and FERET, using metrics like Genuine Acceptance Rate (GAR) versus False Acceptance Rate (FAR), and Equal Error Rate (EER). The results demonstrate superior performance compared to existing methods like Hybrid Approach, BDA, MEB Encoding, Deep CNN, and DH-NND, achieving 100% GAR at 0% FAR with 0% EER for both one-shot and multi-shot enrollment. The system comprises an enrollment phase where facial images are processed to extract, embed, and hash features, and a verification phase where query images are similarly processed for matching against stored templates. The FEHash method significantly enhances face template protection by ensuring that the stored templates do not directly reveal original facial information, thus mitigating risks associated with data breaches and identity theft.","| Database | Enroll. Type | K | GAR@FAR | EER |\n| --- | --- | --- | --- | --- |\n| PIE | One-shot | 256\u003Cbr>1024 | 96.35±0.49%@0.09%\u003Cbr>94.98±0.05%@0.06% | 1.78±0.24%\u003Cbr>2.47±0.02% |\n|  | Multi-shot | 256\u003Cbr>1024 | 96.54±0.35%@0.09%\u003Cbr>95.77±0.22%@0.03% | 1.67±0.18%\u003Cbr>2.09±0.11% |\n| FEI | One-shot | 256\u003Cbr>1024 | 98.76±0.16%@0.01%\u003Cbr>98.25±0.35%@0.006% | 0.61±0.08%\u003Cbr>0.87±0.17% |\n|  | Multi-shot | 256\u003Cbr>1024 | 99.44±0.13%@0.01%\u003Cbr>98.90±0.14%@0.006% | 0.27±0.06%\u003Cbr>0.54±0.07% |\n| FERET | One-shot | 256\u003Cbr>1024 | 97.56±0.36%@0.01%\u003Cbr>96.49±0.59%@0.005% | 1.21±0.18%\u003Cbr>1.74±0.29% |\n|  | Multi-shot | 256\u003Cbr>1024 | 98.11±0.59%@0.01%\u003Cbr>97.48±0.01%@0.005% | 0.93±0.29%\u003Cbr>1.25±0.04% |\n\n\n| Method | Enroll. Type | K | GAR@FAR | EER |\n| --- | --- | --- | --- | --- |\n| Hybrid Approach [12] | Multi-shot | 210 | 90.61%@1% | 6.81% |\n| BDA [11] | Multi-shot | 76 | 96.38%@1% | - |\n| MEB Encoding [30] | Multi-shot | 256\u003Cbr>1024 | 93.22%@0%\u003Cbr>90.13%@0% | 1.39%\u003Cbr>1.14% |\n| Deep CNN [1] | One-shot | 256\u003Cbr>1024 | 91.91%@0.1%\u003Cbr>91.34%@0.1% | 4.00%\u003Cbr>3.60% |\n|  | Multi-shot | 256\u003Cbr>1024 | 97.35%@0%\u003Cbr>96.53%@0% | 0.15%\u003Cbr>0.35% |\n| DH-NND [45] | One-shot | 255\u003Cbr>1023 | 96.2%@0.01%\u003Cbr>96.0%@0.01% | 0.99%\u003Cbr>1.32% |\n|  | Multi-shot | 255\u003Cbr>1023 | 99.9%@0.01%\u003Cbr>99.0%@0.01% | 0.051%\u003Cbr>0.078% |\n| Our Method | One-shot | 256\u003Cbr>1024 | 100%@0%\u003Cbr>100%@0% | 0%\u003Cbr>0% |\n|  | Multi-shot | 256\u003Cbr>1024 | 100%@0%\u003Cbr>100%@0% | 0%\u003Cbr>0% |\n\n| Database | PIE |  | FEI |  | FERET |  |\n| --- | --- | --- | --- | --- | --- | --- |\n| Enroll. Type | One | Multi | One | Multi | One | Multi |\n| D | 5120 | 4096 | 4096 | 4096 | 5120 | 5120 |\n\n\n| NR | 75% | 87.5% | 93.75% | 96.875% | 98.4375% |\n| --- | --- | --- | --- | --- | --- |\n| GAR | 98.94% | 98.31% | 98.25% | 98.13% | 97.25% |\n| FAR | 0.008% | 0.008% | 0.005% | 0.005% | 0.004% |","cbCait429LHhJWUO","https://ap.wps.com/l/cbCait429LHhJWUO","pdf",3006474,10,"English","en",105,"# Full Entropy Hash for Face Template Protection\n## Introduction\n## Related Work\n## Proposed Method\n## Experiments\n## Conclusion","[{\"question\":\"What is the main contribution of this paper?\",\"answer\":\"The main contribution is the proposal of Full Entropy Hash (FEHash), a novel method for enhancing face template protection during biometric recognition.\"},{\"question\":\"How does FEHash ensure template security?\",\"answer\":\"FEHash generates a unique binary string from facial features using random sampling and a hashing function, which is then cryptographically secured for storage, preventing direct reconstruction of the original face.\"},{\"question\":\"What were the key performance metrics used to evaluate FEHash?\",\"answer\":\"The paper evaluated FEHash using metrics such as the Genuine Acceptance Rate (GAR) at a specific False Acceptance Rate (FAR), and the Equal Error Rate (EER) on standard face recognition databases.\"},{\"question\":\"How does FEHash compare to other existing methods?\",\"answer\":\"FEHash demonstrated superior performance, achieving 100% GAR at 0% FAR with 0% EER, outperforming comparative methods like Hybrid Approach, BDA, MEB Encoding, Deep CNN, and DH-NND.\"}]","Dang FEHash-Full Entropy Hash for Face Template Protection | PDF",1788387764,4,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":14,"keywords":34,"description":15,"schema_data":35,"social_meta":90,"head_meta":92,"extra_data":94,"updated_unix":28},"dang-fehash-full-entropy-hash-for-face-template-protection","",{"@graph":36,"@context":89},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":11},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/template/","Template",2,{"item":49,"name":13,"@type":43,"position":50},"https://docshare.wps.com/template/general/",3,{"item":52,"name":14,"@type":43,"position":29},"https://docshare.wps.com/template/dang-fehash-full-entropy-hash-for-face-template-protection/188189/",{"url":52,"name":14,"@type":54,"author":55,"headline":14,"publisher":57,"fileFormat":60,"inLanguage":23,"description":15,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-09-06","2026-09-02",true,{"@type":65,"interactionType":66,"userInteractionCount":11},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81,85],{"name":72,"@type":73,"acceptedAnswer":74},"What is the main contribution of this paper?","Question",{"text":75,"@type":76},"The main contribution is the proposal of Full Entropy Hash (FEHash), a novel method for enhancing face template protection during biometric recognition.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does FEHash ensure template security?",{"text":80,"@type":76},"FEHash generates a unique binary string from facial features using random sampling and a hashing function, which is then cryptographically secured for storage, preventing direct reconstruction of the original face.",{"name":82,"@type":73,"acceptedAnswer":83},"What were the key performance metrics used to evaluate FEHash?",{"text":84,"@type":76},"The paper evaluated FEHash using metrics such as the Genuine Acceptance Rate (GAR) at a specific False Acceptance Rate (FAR), and the Equal Error Rate (EER) on standard face recognition databases.",{"name":86,"@type":73,"acceptedAnswer":87},"How does FEHash compare to other existing methods?",{"text":88,"@type":76},"FEHash demonstrated superior performance, achieving 100% GAR at 0% FAR with 0% EER, outperforming comparative methods like Hybrid Approach, BDA, MEB Encoding, Deep CNN, and DH-NND.","https://schema.org",{"og:url":52,"og:type":91,"og:title":14,"og:site_name":58,"og:description":15},"article",{"robots":93,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":96},[97,102,107,112,117,122,127,132,137],{"id":98,"doc_module":11,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},11,"Presentations",90,"presentations",{"id":103,"doc_module":11,"doc_module_name":46,"category_name":104,"show_sort_weight":105,"slug":106},12,"Resumes",80,"resumes",{"id":108,"doc_module":11,"doc_module_name":46,"category_name":109,"show_sort_weight":110,"slug":111},14,"Invoices",70,"invoices",{"id":113,"doc_module":11,"doc_module_name":46,"category_name":114,"show_sort_weight":115,"slug":116},15,"Posters",60,"posters",{"id":118,"doc_module":11,"doc_module_name":46,"category_name":119,"show_sort_weight":120,"slug":121},16,"Social Media",50,"social-media",{"id":123,"doc_module":11,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},17,"Forms",40,"forms",{"id":128,"doc_module":11,"doc_module_name":46,"category_name":129,"show_sort_weight":130,"slug":131},18,"Letters",30,"letters",{"id":133,"doc_module":11,"doc_module_name":46,"category_name":134,"show_sort_weight":135,"slug":136},21,"Paper Templates",5,"papers-templates",{"id":12,"doc_module":11,"doc_module_name":46,"category_name":13,"show_sort_weight":4,"slug":138},"general-158"]