Comments
Description
Transcript
様々な歩行状況下における歩容認証手法の性能評価
Vol.2013-CVIM-187 No.10 2013/5/30 ใॲཧֶձڀݚใࠂ IPSJ SIG Technical Report ༷ʑͳาߦঢ়گԼʹ͓͚Δา༰ೝূख๏ͷੑೳධՁ ౦ࢁ ါਅ1,a) ᴳ ݪ༃1,b) ߃2,c) ീ ߁࢙1,d) ֓ཁɿۙɼࢦೝূɼ࠼ೝূͳͲʹଓ͘৽͍͠όΠΦϝτϦοΫೝূͱͯ͠ɼηϯα͔ΒΕͨॴ Ͱຊਓ֬ೝ͕Մೳͳา༰ೝূʹߴ͍ؔ৺͕ू·͍ͬͯΔɽ͜Ε·Ͱʹଟ͘ͷา༰ೝূख๏͕ఏҊ͞Εͯ ͍Δ͕ɼͦΕΒͷख๏ʹର͢Δ༷ʑͳঢ়گΛఆͨ͠แׅతͳੑೳධՁະͩͳ͞Ε͍ͯͳ͍ɽͦ͜Ͱɼ ຊͰڀݚɼ༷ʑͳঢ়گԼʹ͓͍ͯɼา༰ೝূख๏ͷੑೳධՁΛߦ͏͜ͱΛతͱ͢ΔɽੑೳධՁʹ༻͍ Δา༰ಛɼγϧΤοτʹͮ͘جͭͷา༰ಛʹՃ͑ͯɼಈ͖ใΛΑΓੵۃతʹར༻ͨ͠า༰ಛ Ͱ͋ΔϑϨʔϜؒࠩʹཱͮ͘جମߴ࣍ݾࣗॴہ૬ؔಛɼΦϓςΟΧϧϑϩʔʹ ͮ͘جGait motion descriptors Λར༻͢Δɽ·ͨɼಛͷর߹ख๏ͱͯ͠ɼϢʔΫϦουͼٴڑਖ਼४ผੳΛ༻͍Δɽ ࣮Ͱݧɼ͕มԽ͢Δ߹ɼ͕มԽ͢Δ߹ͷೋͭͷঢ়͚͓ʹگΔา༰ೝূख๏ͷੑೳධՁΛߦ ͏ɽ࠷ʹޙɼಘΒΕͨ݁ՌΛߟ͠ɼ֤า༰ೝূख๏ͷ༗ޮੑΛൺֱͨ͠ɽ Ωʔϫʔυɿา༰ಛɼੑೳධՁɼมԽɼมԽɼর߹ख๏ 1. ͡Ίʹ ۙɼݸਓใอޢ๏ͷࢪߦతࡒ࢈ͷ૿ՃʹΑΓɼ ϝτϦοΫೝূΛར༻͢ΔϓϩδΣΫτ [1] ͕ਐߦ͍ͯ͠ Δɽ͜ΕɼόΠΦϝτϦοΫೝূͷརͰ͋Δɼޡೝࣝ ͷ͞ɼඃೝূऀʹ͔͔Δෛ୲͕͕ܰ͞ɼΠϯυͷΑ ݸਓೝূγεςϜʹ͓͚ΔηΩϡϦςΟͷڧԽ͕ॏཁࢹ͞ ͏ͳ৽ͯͬͱʹࠃڵड༰͍͢͜͠ͱ͕ཧ༝ͱͯ͛͠ڍΒ Ε͍ͯΔɽ࠷ۙͰɼԕִૢ࡞ΟϧεʹΑΔͷͬ ΕΔɽଞʹɼࢦʹΑΔػࢉܭϞόΠϧͷϩάΠ औΓɼѱ࣭ͳϋοΩϯάʹΑΔݸਓใͷྲྀग़ͱ͍ͬͨ ϯγεςϜɼ੩຺ʹΑΔΞΫηείϯτϩʔϧγεςϜɼ αΠόʔࣄ͕݅ଟൃ͓ͯ͠ΓɼηΩϡϦςΟڴҖ ࢦ DNA ʹΑΔ൜ࡑࠪͷͨΊͷؑఆγεςϜͰར ଞਓࣄͰͳ͍ͱ͜Ζ·Ͱഭ͖͍ͬͯͯΔɽͦͷΑ͏ͳഎ ༻͞Ε͍ͯΔɽ·ͨɼຊࠃʹ͓͍ͯɼۜߦ ATM ʹ ܠͷதɼਓͷੜମใΛར༻ͨ͠όΠΦϝτϦοΫೝূ͕ ͓͍ͯɼ҉ূ൪߸ʹՃ͑ͯࢦͷ੩຺ύλʔϯͷใΛར༻ ߴ͍ؔ৺ΛूΊ͍ͯΔɽ ͢Δ͜ͱͰηΩϡϦςΟͷڧԽΛਤ͍ͬͯΔɽ όΠΦϝτϦοΫೝূͱɼࢦٿ؟ͷ࠼ɼ੩຺ͱ ͔͠͠ͳ͕Βɼࢦɼٿ؟ͷ࠼ɼͱ͍ͬͨମత ͍ͬͨମతಛॺ໊ͱ͍ͬͨߦಈతಛʹΑͬͯຊਓ ಛΛར༻ͨ͠όΠτϝτϦΫεೝূɼηϯα͔Βͷڑ ֬ೝΛߦ͏ೝূํࣜͷ͜ͱͰɼੜମใ͕ͭීวੑɼ། ͕ۙ͘ͳ͚ΕͳΒͳ͍ඃೝূऀ͕ࣗΒొʹ͔ ҰੑɼӬଓੑͱ͍ͬͨಛΛར༻͢Δ͜ͱ͔Βɼ҉ূ൪߸ Θͳ͚ΕͳΒͳ͍ͱ͍͏͕ܽ͋Δɽ͜ͷΑ͏ͳܽΛ ύεϫʔυʹൺݪཧతʹͳݻڧೝূٕज़Ͱ͋Δͱߟ͑ ิ͏όΠΦϝτϦοΫೝূͱͯ͠ɼηϯα͔ΒΕͨҐஔ ΒΕ͍ͯΔɽΞϝϦΧ߹ऺࠃͷύεϙʔτίϯτϩʔϧʹ ͰೝূΛߦ͏͜ͱ͕Ͱ͖Δɼਓͷา͖ํͷݸੑʹͮ͘جา ͓͍ͯɼࢦإͼٴը૾Λऔಘ͢Δ͜ͱͰɼςϩϦετ ༰ೝূ͕ΛूΊ͓ͯΓɼ൜ΧϝϥΛར༻ͨ͠Ҭ ࢦ໊ख൜ͷೖࠃΛ͙ࢼΈ͕ͳ͞Ε͍ͯΔɽΠϯυʹ ࢹ൜ࡑࠪͷԠ༻͕ظ͞Ε͍ͯΔɽ࣮ࡍʹɼΠΪϦ ͓͍ͯɼ12 ԯʹ্ΔେͳͷࠃຽͷݸਓΛೝূ͢ εͰɼڧ౪൜ʹର͢Δา༰ೝূͷ݁Ռ͕ࡋॴʹ͓͚Δ Δख๏ͱͯ͠ɼࢦೝূɼ࠼ೝূɼإೝূͳͲͷόΠΦ ূ͞༻࠾ͯ͠ͱڌΕͨࣄྫ [2] ͕͋Γɼ·ͨɼຊʹ͓͍ ͯɼา༰ೝূʹΑΔؑఆ݁Ռ͕൜ࡑࠪࢧԉʹ͞༻׆Ε 1 2 a) b) c) d) େࡕେֶ Osaka University υϨΫηϧେֶ Drexel University [email protected] [email protected] [email protected] [email protected] ⓒ 2013 Information Processing Society of Japan ͨࣄྫ͋Δɽ ͜ͷΑ͏ͳഎܠͷԼɼ͜Ε·Ͱଟͷา༰ೝূख๏͕ఏ Ҋ͞Ε͖͓ͯͯΓɼͦΕΒେ͖͘ϞσϧϕʔεͱΞϐΞ ϥϯεϕʔεͷ 2 ख๏ʹ͚Δ͜ͱ͕Ͱ͖Δɽ Ϟσϧϕʔεͷख๏Ͱɼೖྗը૾ʹϞσϧΛͯΊΔ 1 Vol.2013-CVIM-187 No.10 2013/5/30 ใॲཧֶձڀݚใࠂ IPSJ SIG Technical Report ͜ͱͰɼਓͷମܕಈ͖ͱ͍ͬͨಛΛநग़͢ΔɽUrtasun Β [3] ɼղֶతͳਓମϞσϧΛͯΊɼؔઅ֯ͷ நग़Λߦ͍ɼSpencer Β [4] ɼநग़͞Εͨؔઅͷपظత ͳҐஔใ͔Βɼ؍ଌํʹରͯ͠ෆมͳؔઅ֯ͷநग़ Λߦͬͨɽ͔͠͠ͳ͕ΒɼϞσϧϕʔεͷख๏ɼղ૾ ͷը૾ʹରͯ͠ɼϞσϧͷͯΊಛநग़ʹࣦഊ͢ (a) GEI (b) FDF (c) GENI Δ߹͕͋Δɽ ΞϐΞϥϯεϕʔεͷख๏Ͱɼը૾Λղੳ͠ಛ Λநग़͢ΔɽHan ͱ Bhanu[5] γϧΤοτը૾ྻΛप ͰظฏۉԽͨ͠ɼGait energy image (GEI) ΛఏҊͨ͠ɽ Makihara Β [6] γϧΤοτը૾Λ্࣌ؒ࣠ʹੵΈॏͶΔ (d) MGEI (e) GFI (f) CGI ͜ͱͰಘΒΕΔา༰γϧΤοτϘϦϡʔϜ͔Βɼ࣌ؒඇґ ଘͷपྖҬಛ (Frequency domain feature, FDF) Λ நग़ͨ͠ɽBashir Β [8] 1 पظͷγϧΤοτը૾ྻ͔ Β࣌ؒతͳΤϯτϩϐʔΛͨ͠ࢉܭɼGait entropy image (GEnI) ΛఏҊͨ͠ɽLam Β [9] γϧΤοτը૾ؒͷΦ ϓςΟΧϧϑϩʔΛͨ͠ࢉܭɼGait flow image (GFI) Λ (g) CHLAC ఏҊͨ͠ɽBashir Β [10] GEnI ͔Β࡞ͨ͠ϚεΫʹ GEI ΛͯΔɼMasked GEI based on GEnI (MGEI) Λఏ Ҋͨ͠ɽWang Β [11] γϧΤοτͷྠֲΛҐ૬ʹ͍ͮج ͯ৭͚ͮͯ͠ 4 ͷ 1 पظຖʹॏͶ߹ΘͤΔ Chrono-gait image (CGI) ΛఏҊͨ͠ɽ͔͠͠ͳ͕ΒɼγϧΤοτʹج (h) GMD า༰ಛ ͮ͘ಛநग़ɼ੩తͳ෦ͱಈతͳ෦Λࠞࡶͯ͠ѻͬ ਤ 1 ͍ͯΔͨΊɼମʹܕେ͖͘ґଘ͢Δ͕͋Δɽ Ұ ํ ɼγ ϧ Τ ο τ ը ૾ Λ ༻ ͍ ͳ ͍ า ༰ ಛ ͱ ͠ ͯ ɼ Kobayashi ͱ Otsu[12] ɼϑϨʔϜؒࠩʹΑΓಈతͳಛ Λऔಘ͠ɼ251 ݸͷۭ࣌ؒύλʔϯΛநग़͢Δɼཱମߴ ख๏ͱͯ͠ɼϢʔΫϦουͮ͘جʹڑর߹ɼਖ਼४ผ ੳʹͮ͘جর߹ͷ 2 ख๏Λ༻͍Δɽ ࣍ݾࣗॴہ૬ؔ (Cubic higher-order local auto-corelation, 2. า༰ಛ CHLAC) ΛఏҊͨ͠ɽ·ͨɼBashir Β [13] ɼೖྗը૾ؒ 2.1 γϧΤοτʹͮ͘جา༰ಛ ͷΦϓςΟΧϧϑϩʔΛ͠ࢉܭɼಛΛ্Լࠨӈͷಈɾ ΞϐΞϥϯεϕʔεͷา༰ಛͷதͰɼา༰γϧΤο ੩ࢭͷͭޒͷهड़ࢠʹผ͢ΔɼGait motion descriptor τը૾͔Βಛநग़Λߦ͏ํ๏͕ࡏݱͷओྲྀͱͳ͍ͬͯ (GMD) ΛఏҊͨ͠ɽ Δɽา༰γϧΤοτը૾ɼഎࠩܠʹͮ͘جάϥϑΧο ͜ͷΑ͏ʹ༷ʑͳา༰ೝূख๏͕ఏҊ͞ΕΔதɼIwama Β [14] ΞϐΞϥϯεϕʔεͷา༰ೝূख๏ʹ͍ͭͯɼ τྖҬׂ [7] ͳͲΛ༻͍ͯ࡞͞ΕΔɽ ·ͨɼา༰ೝূಈը૾Λରͱ͢ΔͷͰɼর߹ͷࡍʹ 4,000 ਓҎ্ͷେنσʔλϕʔεΛ༻͍ͨੑೳධՁΛ ಉҰ࢟ಉ࢜ͷ੩ࢭըͰରԠͤ͞ΔͨΊͷϑϨʔϜಉظ ߦͬͨɽ͔͠͠ͳ͕Βɼ্هͷੑೳධՁࡾͭͷݶͰ؍ ॲཧ͕ඞཁͰ͋Δɽͦ͜Ͱɼ1 पظͷಈը૾͔Β։࢝ϑ ఆతͰ͋Δͱ͍͏͕͋ͬͨɽୈҰʹɼಉঢ়گԼͰͷ ϨʔϜͷาߦ࢟ʹґΒͳ͍า༰ಛΛऔಘ͠ɼͦͷपظ ੑೳධՁ͔͠ߦΘΕ͍ͯͳ͍͜ͱɼୈೋʹɼγϧΤοτ ୯Ґͷಛຖʹর߹Λߦ͏͜ͱͰϑϨʔϜಉॲظཧΛ؆୯ ը૾ʹͮ͘جͭͷา༰ಛ (GEIɾFDFɾGEnIɾGFIɾ Խ͢Δɽपݕظग़ͷख๏ʹ͍ͭͯɼ[6] ͰఏҊ͞Ε͍ͯΔ MGEIɾCGI) ʹ͍ͭͯͷΈͷੑೳධՁʹݶఆ͞Ε͍ͯΔ γϧΤοτը૾ྻͷਖ਼نԽࣗݾ૬ؔ࠷େԽʹͮ͘جख๏Λ ͜ͱɼୈࡾʹɼϢʔΫϦουͮ͘جʹڑ୯Ұͷর߹ख๏ ༻͍ΔɽҎ߱ɼຊ͍༻ͰڀݚΔา༰ಛʹ͍ͭͯɼ؆୯ʹ ʹݶఆ͞Ε͍ͯΔ͜ͱ͕͛ڍΒΕΔɽ આ໌͓ͯ͘͠ɽ ΑͬͯຊͰڀݚɼҎԼͷࡾʹΑΓɼΑΓแׅతͳา GEI ɼγϧΤοτը૾ྻΛาߦपͰظฏۉԽ͢Δ͜ͱ ༰ೝূख๏ͷੑೳධՁΛߦ͏͜ͱΛతͱ͢Δɽ(1) ͰಘΒΕΔɽGEI ਓྖҬ෦ɼͭ·Γମܕͷେ͖͕͞ มԽมԽͱ͍༷ͬͨʑͳঢ়گԼʹ͓͍ͯੑೳධՁΛ ಛʹେ͖͘࡞༻͢ΔͷͰɼಉঢ়گԼͰͷর߹ʹ͓͍ͯߴ ߦ͏ɽ(2) γϧΤοτʹͮ͘جา༰ಛʹՃ͑ɼϑϨʔϜ ͍ੑೳΛൃ͢شΔ [14]ɽ͔͠͠ͳ͕ΒɼมԽͷΑ͏ͳ ؒࠩɾΦϓςΟΧϧϑϩʔͱ͍ͬͨɼ੩ࢭ෦ͱಈ͖෦ ਓྖҬͷγϧΤοτ͕େ͖͘มԽ͢Δ߹ʹਫ਼͕ Λ۠ผͨ͠า༰ಛʹ͍ͭͯੑೳධՁΛߦ͏ɽ(3) র߹ ͘ͳΔ͕͋Δɽਤ 1(a) GEI ΛՄࢹԽͨ͠ը૾Ͱ ⓒ 2013 Information Processing Society of Japan 2 Vol.2013-CVIM-187 No.10 2013/5/30 ใॲཧֶձڀݚใࠂ IPSJ SIG Technical Report ͋Δɽ Ϋηϧʹ͍ͭͯɼۭ࣌ؒతͳߴ࣍ॴہύλʔϯͱͷ૬ؔΛ FDF ɼาߦपͮ͘جʹظपྖҬಛΛา༰ಛ ͠ࢉܭɼ֤ύλʔϯʹର͢ΔώετάϥϜΛ࡞͠ɼͦΕ ͱͯ͠ར༻͢Δɽपظຖʹׂͨ͠γϧΤοτը૾ྻʹର Λಛͱ͢ΔɽCHLAC ݾࣗॴہ૬ؔಛͷੑ࣭Ͱ͋Δ ͯ͠ɼ࣌ؒ࣠ํͷ 1 ࣍ࢄݩϑʔϦΤมΛ͠ࢉܭɼา Ճ๏ੑ͕͋Δ͜ͱ͔Βɼর߹ͷࡍʹಈը૾͔ΒਓྖҬΛ ߦपͰظਖ਼نԽ͞Εͨपͷৼ෯εϖΫτϧΛநग़ நग़͢Δඞཁ͕ͳ͍ɽ·ͨɼଞͷา༰ಛʹൺͯ࣍ݩ ͢Δ͜ͱͰಘΒΕΔɽFDF GEI ಉ༷ɼਓྖҬ͕ಛ ͷಛऔಘ͢Δ͜ͱ͕Ͱ͖Δɽਤ 1(g) CHLAC ͷώε ʹେ͖͘࡞༻͢Δɽಛʹ 0 ഒप GEI ͱಉ༷ͷܭ τάϥϜͰ͋Δɽ ࢉࣜʹͳΔͨΊɼGEI ͱಉ༷ͷಛ͕நग़͞ΕΔɽਤ 1(b) FDF ΛՄࢹԽͨ͠ը૾Ͱ͋Δɽ GEnI ɼGEI ʹର͢ΔγϟϊϯΤϯτϩϐʔΛࢉग़͢ 2.3 ΦϓςΟΧϧϑϩʔʹͮ͘جา༰ಛ าߦʹ͓͚Δಈతͳ෦Λநग़͢Δख๏ͱͯ͠ɼΦϓ Δ͜ͱͰಘΒΕΔɽGEnI ମװ෦ͳͲͷपͰظલܠ ςΟΧϧϑϩʔʹͮ͘جख๏͕͋Γɼͦͷදྫͱͯ͠ɼ Ͱ͋Γଓ͚ΔྖҬɼഎ͋ͰܠΓଓ͚ΔྖҬɼଈͪɼGEI GMD ͕͋Δɽ ʹ͓͍ͯനࠇͱͳΔྖҬʹ͓͚Δ 0 ͱͳΓɼखͳ GFI ͷΑ͏ʹγϧΤοτը૾ʹ͢ࢉܭ͍ͯͮجΔΦϓ Ͳͷಈ͖ͷ͋ΔྖҬɼଈͪɼGEI ʹ͓͍ͯփ৭ͱͳΔྖҬ ςΟΧϧϑϩʔͱҟͳΓɼGMD ମ෦ͷςΫενϟ Ͱେ͖ͳͱͳΔΑ͏ͳಛͰ͋ΔɽΑͬͯ GEI ͱൺ ༻͍Δ͜ͱ͔ΒɼΑΓ๛Ͱ࣮ࡍతͳಈ͖ಛΛಘΔ͜ ͯɼΑΓಈ͖Λॏࢹͨ͠ಛΛநग़͢Δ͜ͱ͕Ͱ͖Δɽͦ ͱ͕Ͱ͖Δɽ·ͨɼGMD ͰγϧΤοτը૾ΛϚεΫͱ ͷੑ࣭͔ΒɼมԽͷΑ͏ͳਓྖҬ͕େ͖͘มԽ͢Δ ͯ͠༻͍Δ͜ͱͰࠨӈ্Լํͷಈ͖෦ͱɼͦΕҎ֎ͷ ঢ়Ͱگൺֱతর߹͕͏·͍͘͘Մೳੑ͕͋Δɽਤ 1(c) ੩ࢭ෦Λͯ͠औಘ͢Δ͜ͱ͕ՄೳͰ͋Δɽਤ 1(h) GEnI ΛՄࢹԽͨ͠ը૾Ͱ͋Δɽ MGEI ্هͷ GEIɼGEnI ΛΈ߹Θͤͨา༰ಛͰ GMD ͷ੩ࢭ෦ɼࠨɾӈɾ্ɾԼํͷಈ͖෦ΛՄࢹ Խͨ͠ը૾Ͱ͋Δɽ ͋ΔɽGEnI Ͱେ͖ͳΛͭྖҬɼଈͪɼಈ͖ͷେ͖ͳ GMD ɼา༰ಛΛ੩తͳ෦ͱಈతͳ෦ʹ۠ผ͢ ྖҬΛϚεΫͱͯ͠औΓग़͠ɼͦͷϚεΫΛ GEI ʹద༻ ΔͨΊɼมԽมԽͳͲɼ༷ʑͳঢ়گมԽʹର͠ ͢Δ͜ͱͰಛΛநग़͢ΔɽMGEI ಈ͖Λॏࢹͨ͠ಛ ͯͳͱ݈ؤΔՄೳੑ͕͋Δɽর߹ͷࡍɼࠨӈ্Լͷಈ͖ Ͱ͋Γɼבίʔτͱ͍ͬͨਓྖҬͷมԽʹର݈ͯ͠ؤ ํͱ੩ࢭ෦ͷͭޒͷهड़ࢠຖʹ૬ҧΛٻΊΔɽ͜͜ ͳʹ͋Δɽਤ 1(d) GEI ΛՄࢹԽͨ͠ը૾ͱ GEnI ͰɼٻΊͨͭޒͷ૬ҧͷɼԼํͷهड़ࢠࣝผʹ༗ ͔Β࡞ͨ͠ϚεΫը૾Ͱ͋Δɽ ޮͳใྔ͕গͳ͍ͨΊল͖ɼΓͷ࢛ͭͷ૬ҧͷॏΈ GFI ɼγϧΤοτը૾ྻʹର͢ΔΦϓςΟΧϧϑϩʔ ͔ΒɼᮢҎ্ͷಈ͖ͷ͋Δ෦ΛपͰظฏۉԽ͢Δ͜ ͚૯Λ࠷ऴతͳ૬ҧͱ͢Δɽ ͱͰநग़͢Δา༰ಛͰ͋ΔɽGFI ΦϓςΟΧϧϑϩʔ 3. র߹ख๏ ۭ͔ؒΒநग़͞ΕΔͨΊɼγϧΤοτը૾ྻͷಈ͖ͷ෦ 3.1 ϢʔΫϦουͮ͘جʹڑর߹ख๏ Λॏࢹͨ͠า༰ಛͱͳΔɽਤ 1(e) GFI ΛՄࢹԽͨ͠ র߹ख๏ͱͯ͠࠷୯७ͳํ๏ɼา༰ಛͷ֤࣍ݩͷ ը૾Ͱ͋ΔɽγϧΤοτը૾ʹର͢ΔΦϓςΟΧϧϑϩʔ ࠩͷࣗͷฏํࠜɼଈͪɼา༰ಛϕΫτϧʹର͢ΔϢʔ Λ͍ͯ͠ࢉܭΔͷͰɼಛ͕ਓͷྠֲۙʹͷΈݱΕͯ ΫϦουڑΛ૬ҧͱ͢Δํ๏Ͱ͋Δɽ͜͜Ͱɼา༰ ͍Δ͜ͱ͕͔Δɽ ը૾ྻ P ɼG ͔Βநग़͞ΕΔ D ࣍ݩͷา༰ಛϕΫτϧΛ CGI ɼ4 ͷ 1 าߦपظຖʹநग़ͨ͠ྠֲΛҐ૬ʹԠ D P G D G xP i ∈ R (i = 1, . . . , N )ɼ ͼٴx j ∈ R (j = 1, . . . , N ) ٖͯ͡ࣅ৭Λ༻͍ͯϚοϐϯά͠ɼॏͶ߹Θͤͨา༰ಛ Λর߹͢Δ͜ͱΛߟ͑Δɽ͜͜ͰɼN P ɼN G ɼา༰ը Ͱ͋ΔɽCGI ྠֲΛಛͱͯ͠நग़͍ͯ͠Δ͜ͱ͔Βɼ ૾ྻ P ɼG ͔ΒͦΕͧΕநग़͞ΕΔา༰ಛ (ຆͲͷಛ มԽͳͲ֎తཁҼʹΑΔาߦঢ়گͷมԽʹͳ݈ؤ ʹ͓͍ͯपظͷ) Ͱ͋Δɽ·ͣɼP ͷ i ൪ͷಛ ʹ͋Δɽਤ 1(f) CGI ΛՄࢹԽͨ͠ը૾Ͱ͋Δɽ G xP i ͱɼG ͷ j ൪ͷಛ x j ͷ૬ҧ Mi,j ΛҎԼͷΑ͏ ʹ͢ࢉܭΔɽ 2.2 ϑϨʔϜؒࠩʹͮ͘جา༰ಛ าߦʹ͓͚Δಈతͳ෦Λநग़͢Δख๏ͱͯ͠ɼϑϨʔ Ϝ͕ؒࠩ͋ΔɽϑϨʔϜؒࠩɼt ϑϨʔϜͷը૾ ͱ t + n ϑϨʔϜͷը૾ͷ͕ࠩᮢҎ্ͷͱͯ͠ந ग़͞ΕΔɽ ϑϨʔϜؒࠩͰ࡞ͨ͠ಈͷγʔέϯε͔ΒಛΛ நग़͢Δख๏ͱͯ͠ɼCHLAC ͕͋ΔɽCHLAC Ͱɼϑ ϨʔϜؒࠩʹͯநग़ͨ͠ಈ͖ྖҬʹ͓͚ΔͦΕͧΕͷϐ ⓒ 2013 Information Processing Society of Japan G Mi,j = x P i − xj (1) ࠷ऴతͳ૬ҧ M ɼ֤ಛͷʹର͢Δ૬ҧ Mi,j ͷ ࠷খΛબ͢Δɽ M = min Mi,j i,j (2) 3.2 ਖ਼४ผੳʹͮ͘جর߹ख๏ ೝূੑೳͷ্ͷͨΊʹɼຊਓಉ࢜ͷڑΛ࠷খԽ͠ɼ 3 Vol.2013-CVIM-187 No.10 2013/5/30 ใॲཧֶձڀݚใࠂ IPSJ SIG Technical Report ද 1 OU-ISIR Gait Database, The Treadmill Dataset આ໌ ֶश ςετ σʔληοτ (a) PCA Dataset A 9 छྨͷมԽ 14 ਓ 20 ਓ Dataset B ࠷େ 32 छྨͷมԽ 21 ਓ 47 ਓ (b) LDA ਤ 2 ֤ۭؒʹ͓͚Δ GEI ͷࢄਤ ଞਓಉ࢜ͷڑΛ࠷େԽ͢ΔΑ͏ͳಛۭؒΛߏங͢Δ͜ ͱ͕ॏཁͰ͋ΔɽͦͷΑ͏ͳۭؒઢܗผੳ (Linear discriminant analysis, LDA) Λར༻͢Δ͜ͱͰಘΒΕΔɽ ઢܗผੳͱɼΫϥεڞࢄߦྻͱΫϥεؒڞࢄ ߦྻͷҰൠԽݻ༗Λղ͘͜ͱͰɼΫϥεࢄΛ࠷ ਤ 3 มԽ (্ஈ) ͱมԽ (Լஈ) ͷྫ খԽ͠ɼΫϥεؒࢄΛ࠷େԽ͢ΔΑ͏ͳಛۭؒͷجఈ ϕΫτϧΛಘΔख๏Ͱ͋Δɽ͜͜Ͱɼߴ࣍ݩͷา༰ಛ Λओੳ (Principal component analysis, PCA) Λ༻ 4. ࣮ݧ ͍ͯ࣍ݩͷۭؒʹࣹӨ͠ɼಘΒΕͨ࣍ݩσʔλΛઢܗ 4.1 σʔληοτ ผੳ (LDA) ͢Δɼਖ਼४ผੳΛ༻͍Δɽ ࣮ʹݧ The OU-ISIR Gait Database, The Treadmill ·ͣɼֶशʹ༻͍ΔಛͷฏۉϕΫτϧΛ༻͍ͯɼ֤ಛ Dataset [15] ͷɼDataset AɼB Λ༻͍ͨɽͳ͓ɼDataset ͔Βࠩ͠Ҿ͍͓͖ͯɼओੳΛ༻͍ͯา༰ಛΛ A 2km/h ͔Β 10km/h ·Ͱ 1km/h ࠁΈͷมԽΛ࣋ ࣹ࣍ʹۭؒݩӨ͢Δɽ࣍ʹɼֶशʹ༻͍Δಛɼಛͷ ͭ 34 ਓͷඃऀݧͷาߦγʔέϯε͔ΒͳΓɼDataset B ࣍ݩΛͦΕͧΕߦɼྻͱͨ͠ߦྻʹରͯ͠ಛҟղΛద ࠷େ 32 छྨͷมԽΛ࣋ͭ 68 ਓͷඃऀݧͷาߦγʔέ ༻͢Δ͜ͱͰɼ࣍ݩͷۭؒͷجఈϕΫτϧΛٻΊΔɽਤ ϯε͔ΒͳΔɽ 2(a) GEI ͷ PCA ۭؒʹ͓͚Δ্Ґ 2 ओͷࢄਤͰ ͋Δɽ ද 1 ʹ༻ͨ͠าߦγʔέϯεΛవΊΔɽ·ͨɼม ԽɼมԽͷྫΛਤ 3 ʹࣔ͢ɽ ࣍ʹɼಘΒΕͨ࣍ݩσʔλΛઢܗผੳ͢Δɽઢܗ ·ͨɼา༰ಛͷর߹ʹࡍͯ͠ɼσʔληοτΛΪϟϥ ผੳͰɼԼهͷ௨Γɼ࣍ݩσʔλͷΫϥεڞ Ϧʔͱϓϩʔϒͷೋछྨʹׂ͢ΔɽΪϟϥϦʔาߦ ࢄ SW ͱΫϥεؒڞࢄ SB Λ͠ࢉܭɼೋͭͷڞࢄߦྻ γʔέϯεͷొσʔλΛɼϓϩʔϒาߦγʔέϯεͷ ͷҰൠԽݻ༗Λղ͘ɽ ೖྗσʔλΛҙຯ͢Δɽ͜͜ͰɼDatasetA ඃऀݧຖ SW = Nfi eatures N class i=1 ʹಉঢ়گԼͰͷาߦγʔέϯε͕ 2 ύλʔϯ͋ΔͷͰɼͦ (xi,j − mi )(xi,j − mi )T (3) j=1 ΕͧΕΛΪϟϥϦʔɼϓϩʔϒͱ͠ɼDataset B ܰʹ ͋ͨΔ 1 छྨͷาߦγʔέϯεΛΪϟϥϦʔɼΓͷ࠷େ 31 छྨͷͷาߦγʔέϯεΛϓϩʔϒͱ͢Δɽ SB = N class Nfi eatures (mi − m)(mi − m)T (4) i=1 SW x = λSB x, x = 0 ͜͜ͰɼNclass ඃऀݧΛɼNfi eatures (5) 4.2 ධՁई า༰ೝূͷੑೳʹ͓͍ͯɼ1 ର 1 ೝূ (Vefirication) ͱ 1 ର N ೝূ (Identification) ͷೋͭͷγφϦΦΛߟྀ͢Δɽ i ൪ͷΫϥε 1 ର 1 ೝূͰɼর߹ʹΑΓಘΒΕͨ૬ҧ͕ᮢҎԼ ͷಛΛද͢ɽ·ͨɼxi,j i ൪ͷΫϥεͷ j ൪ͷ Ͱ͋Εຊਓͱͯ͠ड͚ೖΕɼᮢΛ͑Εଞਓͱ͠ ಛΛɼmi i ൪ͷΫϥεͷฏۉΛɼm શಛͷฏ ͯڋ൱͢ΔɽϓϩʔϒͱΪϟϥϦʔશͯͷʹରͯ͜͠ ۉΛද͢ɽ ͷΑ͏ͳஅΛߦ͍ɼ݁Ռͱͯ͠ຊਓڋ൱ޡΓ (False ਤ 2(b) GEI LDA ۭؒʹ͓͚Δ্Ґ 2 ͷࢄਤ rejection rate, FRR) ͱଞਓडೖޡΓ (False acceptance Ͱ͋ΔɽPCA ۭؒʹ͓͚Δࢄਤ (ਤ 2(a)) ʹൺɼLDA rate, FAR) ΛٻΊΔɽຊਓڋ൱ޡΓͱଞਓडೖޡΓͷ ۭؒͷࢄਤ (ਤ 2(b)) ֤ΫϥεͰີू͠ɼ֤Ϋϥε Έ߹Θͤɼ૬ҧͷᮢΛมԽͤ͞Δ͜ͱͰมԽ͢Δ͜ ؒͰࢄ͍ͯ͠Δ͜ͱ͕͔Δɽ ͱ͔ΒɼͦͷτϨʔυΦϑΛද͢ड৴ऀૢ࡞ಛੑ (Receiver Operating Characteristics, ROC) ۂઢ͕ಘΒΕΔɽ ⓒ 2013 Information Processing Society of Japan 4 Vol.2013-CVIM-187 No.10 2013/5/30 ใॲཧֶձڀݚใࠂ IPSJ SIG Technical Report ͜͜Ͱɼਤ 4(a), (b) ϢʔΫϦουͮ͘جʹڑর߹ख๏ ʹର͢Δ ROC ۂઢͱ CMC ۂઢΛɼਤ 4(c), (d) ਖ਼४ ผੳʹͮ͘جর߹ख๏ʹର͢Δ ROC ۂઢͱ CMC ۂઢ Λ͍ࣔͯ͠Δɽ·ͨɼਤ 4(e)ɼ(f) ɼͦΕͧΕՁޡΓ ɼ1 Ґೝূͷ݁ՌΛ͍ࣔͯ͠Δɽ มԽʹ͓͚Δา༰ೝূͷ߹ɼর߹͢ΔΪϟϥϦʔ ͱϓϩʔϒͷਓྖҬͷγϧΤοτ͕େ͖͘มԽ͢Δͨ ΊɼγϧΤοτʹͮ͘جา༰ಛͷਫ਼͕͘ͳΔɽಛʹ (a) ROC ۂઢ (ϢʔΫϦου (b) CMC ۂઢ (ϢʔΫϦου ਓྖҬͷγϧΤοτʹେ͖͘ґଘ͢Δ GEI FDF ଞ )ڑ )ڑ ͷา༰ಛͱൺͯೝূਫ਼͕͍ɽҰํɼΤϯτϩϐʔ Λར༻ͨ͠ GEnI MGEI γϧΤοτʹͮ͘جา༰ಛ ͷதͰಈ͖෦ʹண͍ͯ͠ΔͨΊɼߴ͍ਫ਼ͱͳͬͯ ͍Δ (ਤ 4(a), (b) ࢀর)ɽ ͔͠͠ͳ͕Βɼਖ਼४ผੳʹͮ͘جর߹ͷ݁Ռ (ਤ 4(c), (d)) ΛݟΔͱɼGEI FDF ͕ GEnI MGEI ΑΓߴ͍ਫ਼ ͕ग़͍ͯΔɽ͜Ε GEnI MGEI ͕୯७ൺֱͷࡍʹਫ਼ ্͕͢ΔͷͰ͋ͬͯɼਖ਼४ผੳʹͮ͘جর߹ख (c) ROC ۂઢ (ਖ਼४ผੳ) (d) CMC ۂઢ (ਖ਼४ผੳ) ๏ͷ߹ɼGEI FDF ͷํ͕ΑΓઢܗผੳ͕ޮՌత ʹಇۭؒ͘Ͱ͋ͬͨ͜ͱ͕ཧ༝ͱͯ͛͠ڍΒΕΔɽ ·ͨਤ 4(e), (f) ͷ݁ՌΑΓɼਖ਼४ผੳʹͮ͘جর߹ ख๏͕ϢʔΫϦουͮ͘جʹڑর߹ख๏ΑΓ༏Ε͍ͯΔ ͜ͱ͕͔ΔɽΑͬͯɼ࣍અҎ߱ਖ਼४ผੳΛ༻͍ͨ ݁ՌͷΈΛఏࣔ͢Δɽ 4.4 มԽʹ͓͚ΔੑೳධՁ (e) ՁޡΓ ਤ 4 (f) 1 Ґೝূ มԽʹ͓͚ΔੑೳධՁ ࣍ʹɼมԽʹ͓͚Δา༰ೝূͷੑೳධՁΛߦ͏ɽ͜ ͜Ͱɼਤ 5 1 ର 1 ೝূͷ݁ՌΛɼਤ 6 1 ର N ೝূͷ݁ ՌΛ͍ࣔͯ͠Δɽ·ͨɼਤ 7(a)ɼਤ 7(b) ͦΕͧΕՁ 1 ର 1 ೝূʹ͓͍ͯɼຊਓڋ൱ޡΓͱଞਓडೖޡΓ ޡΓɼ1 Ґೝূͷ݁ՌΛ͍ࣔͯ͠Δɽ ͷ྆ํ͕͍͜ͱ͕·͍͠ͱ͑ݴΔɽ·ͨɼຊਓڋ൱ มԽʹ͓͚Δา༰ೝূͷ߹ɼর߹͢ΔΪϟϥϦʔ ޡΓͱଞਓडೖޡΓ͕͍͠ͱ͖ɼͦͷޡΓΛՁ ͱϓϩʔϒͷਓͷಈ͖͕େ͖͘มԽ͢ΔͨΊɼಈ͖෦ ޡΓ (Equal error rate, EER) ͱͼݺɼ1 ର 1 ೝূʹ͓͚ ʹணͨ͠า༰ಛͷਫ਼͘ͳΔɽಛʹΦϓςΟΧϧ Δࢦඪͱ͢Δɽ ϑϩʔΛར༻ͨ͠ GFI GMD ΑΓಈ͖෦Λར༻͢ 1 ର N ೝূͰɼ͋Δೖྗಛͱ N ਓͷొಛͷর ΔͨΊɼଞͷา༰ಛͱൺͯࠩʹΑΔਫ਼ͷԼ ߹ͰಘΒΕͨ N ݸͷ૬ҧΛʹݩϥϯΩϯάΛ࡞͢Δɽ ͕େ͖͍ɽҰํɼਓͷܗঢ়ʹେ͖͘ґଘ͢Δ GEI ͜ͷͱ͖ɼ૬ҧ͕͍΄ͲϥϯΩϯάͰ্ҐͱͳΔɽ࣍ FDF ɼಈ͖ಛͱൺֱ͢Δͱਫ਼ͷԼ͕ݶఆతͱͳͬ ʹɼ࡞ͨ͠ϥϯΩϯά͔Β֤ϓϩʔϒʹ·ؚΕΔຊਓಉ ͍ͯΔɽ ࢜ͷॱҐΛௐɼͦͷॱҐʹର͢ΔྦྷੵΛྦྷੵࣝผਫ਼ 5. ߟ ಛੑ (Cumulative matching characteristics, CMC) ۂઢ ͱͯ͠ࢉग़͢ΔɽCMC ۂઢʹ͓͍ͯɼྫ͑ɼ3 Ґʹର ༷ʑͳาߦঢ়گԼʹ͓͚Δา༰ೝূख๏ͷੑೳධՁΛ ͢Δೝূ͕ 90%Ͱ͋Εɼຊਓಉ࢜ͷর߹ϖΞ͕ 3 ҐҎ ߦͬͨ݁Ռɼর߹ख๏ͰϢʔΫϦουʹڑൺͯਖ਼४ ʹೖΔׂ߹͕ 90%Ͱ͋Δͱ͍͏͜ͱΛҙຯ͍ͯ͠Δɽ ผੳΛ༻͍Δͷͷํ͕ੑೳ͕ྑ͍͜ͱ͕͔ͬͨɽ า༰ೝূͷඪͱͯ͠ɼॱҐ͕͍ͱ͖ͷೝূ͕ߴ͍ ͜Εਖ਼४ผੳʹΑΔΫϥεؒࢄΛ࠷େԽɼΫϥε ͜ͱ͕·͘͠ɼಛʹ 1 Ґೝূ͕ੑೳධՁʹ͓͍ͯྑ͘ ࢄΛ࠷খԽ͢Δۭؒͷߏங͕ཧ༝ͱͯ͛͠ڍΒΕΔɽ ༻͍ΒΕΔɽ Ұํɼา༰ಛʹ͍ͭͯߟ͑ͯΈΔͱɼาߦঢ়ʹگΑͬ ͯਫ਼ͷେ͖ͳҧ͍͕ݟΒΕͨɽ͜͜ͰγϧΤοτʹج 4.3 มԽʹ͓͚ΔੑೳධՁ ࠷ॳʹɼมԽʹ͓͚Δา༰ೝূͷੑೳධՁΛߦ͏ɽ ⓒ 2013 Information Processing Society of Japan ͮ͘า༰ಛͰ͋Δ GEI ͱಈ͖Λ໌ࣔతʹར༻ͨ͠า༰ ಛͰ͋Δ GMD ΛྫʹऔΓ্͛ͯɼߟ͢Δɽ 5 Vol.2013-CVIM-187 No.10 2013/5/30 ใॲཧֶձڀݚใࠂ IPSJ SIG Technical Report (a) 2 km/h vs. 4 km/h (b) 2 km/h vs. 6 km/h (a) 2 km/h vs. 4 km/h (b) 2 km/h vs. 6 km/h (c) 2 km/h vs. 8 km/h (d) 2 km/h vs. 10 km/h (c) 2 km/h vs. 8 km/h (d) 2 km/h vs. 10 km/h (e) 4 km/h vs. 6 km/h (f) 4 km/h vs. 8 km/h (e) 4 km/h vs. 6 km/h (f) 4 km/h vs. 8 km/h (g) 4 km/h vs. 10 km/h (h) 6 km/h vs. 8 km/h (g) 4 km/h vs. 10 km/h (h) 6 km/h vs. 8 km/h (i) 6 km/h vs. 10 km/h (j) 8 km/h vs. 10 km/h (i) 6 km/h vs. 10 km/h (j) 8 km/h vs. 10 km/h ਤ 5 มԽʹ͓͚Δ ROC ۂઢʹΑΔੑೳධՁ ·ͣɼมԽʹ͍ͭͯߟ͢Δɽ͕มԽ͢Δ߹ɼ ਤ 6 มԽʹ͓͚Δ CMC ۂઢʹΑΔੑೳධՁ ͯ GEI Λ, ಈ͖Λ໌ࣔతʹར༻ͨ͠า༰ಛͱͯ͠ GMD લड़ͷ௨ΓɼਓͷγϧΤοτܗঢ়͕େ͖͘มԽ͢ΔͷͰ Λͨ͠ࢉܭͷ͕ɼਤ 8(b)ɼ8(c)ɼ8(e)ɼ8(f) Ͱ͋Δɽ·ͨɽ า༰ͷಈ͖ʹண͢Δ͜ͱ͕ॏཁͰ͋Δɽਤ 8(a)ɼਤ 8(d) ͜ΕΒͷը૾ؒͷࠩΛͨ͠ࢉܭͷ͕ɼਤ 8(g)ɼ8(h) Ͱ ಉҰਓʹΑΔҟͳΔͷาߦը૾Ͱ͋Δɽ͜ΕΒͷ ͋Δɽ͜ͷͱ͖ɼGEIɼGMD ʹ͓͚Δ੩ࢭ෦Ͱ 2 ಛ าߦγʔέϯεʹ͍ͭͯγϧΤοτʹͮ͘جา༰ಛͱ͠ ؒʹେ͖ͳ͕ࠩݟΒΕΔ͕ɼGMD ʹ͓͚Δಈ͖෦Ͱ ⓒ 2013 Information Processing Society of Japan 6 Vol.2013-CVIM-187 No.10 2013/5/30 ใॲཧֶձڀݚใࠂ IPSJ SIG Technical Report (a) ՁޡΓ ਤ 7 (a) (a) (b) (c) (d) (e) (f) (b) 1 Ґೝূ มԽʹ͓͚ΔՁޡΓͱ 1 Ґೝূ (b) (c) (g) (h) ਤ 9 มԽʹ͓͚Δา༰ಛͷҧ͍ɽ্ஈ: ೖྗಛɼதஈ: ొ ಛɼԼஈ: ೖྗಛͱొಛͷࠩɽ্ஈɾதஈʹ͓͍ͯ ɼࠨ͔Βɼݪը૾ɼGEIɼGMD Λද͢ɽ·ͨɼ(g)ɼ(h) ɼ ೖྗͱొͷ GEI ͱ GFI ͷࠩΛද͢ɽ͜͜Ͱɼೖྗಛ (d) (e) (f) ͷΈʹଘࡏ͢Δ෦ɼొಛͷΈʹଘࡏ͢Δ෦Λ ද͠ɼͦΕͧΕൺֱରͷา༰ಛྖҬΛද͢ɽͭ·Γɼɼ ͘͠Ͱද͞Ε͍ͯΔྖҬൺֱ͢ΔରͱͷࠩΛࣔ͢ɽ ʹٯແ࠼৭ͱͳ͍ͬͯΔྖҬɼํͰಛ͕͍͠෦Ͱ ͋Δɽ έϯεʹ͍ͭͯɼγϧΤοτʹͮ͘جา༰ಛͱͯ͠ GEI Λɼಈ͖Λ໌ࣔతʹར༻ͨ͠า༰ಛͱͯ͠ GMD Λࢉܭ (g) ਤ 8 (h) ͨ͠ͷ͕ɼਤ 9(b)ɼ9(c)ɼ9(e)ɼ9(f) Ͱ͋Δɽ·ͨɽ͜ มԽʹ͓͚Δา༰ಛͷҧ͍ɽ্ஈ: ೖྗಛɼதஈ: ొ ΕΒͷը૾ؒͷࠩΛͨ͠ࢉܭͷ͕ɼਤ 9(g)ɼ9(h) Ͱ͋ ಛɼԼஈ: ೖྗಛͱొಛͷࠩɽ্ஈɾதஈʹ͓͍ͯ Δɽ͜ͷͱ͖ɼGMD ʹ͓͚Δಈ͖෦Ͱ 2 ಛؒʹେ ɼࠨ͔Βɼݪը૾ɼGEIɼGMD Λද͢ɽ·ͨɼ(g)ɼ(h) ɼ ͖ͳ͕ࠩݟΒΕΔ͕ɼGEI GMD ʹ͓͚Δ੩ࢭ෦Ͱ ೖྗͱొͷ GEI ͱ GFI ͷࠩΛද͢ɽ͜͜Ͱɼೖྗಛ ͕ࠩখ͍͞ɽ͜ͷ݁Ռ͔ΒมԽʹ͓͍ͯɼਓͷγϧ ͷΈʹଘࡏ͢Δ෦ɼొಛͷΈʹଘࡏ͢Δ෦Λ ද͠ɼͦΕͧΕൺֱରͷา༰ಛྖҬΛද͢ɽͭ·Γɼɼ ͘͠Ͱද͞Ε͍ͯΔྖҬൺֱ͢ΔରͱͷࠩΛࣔ͢ɽ ʹٯແ࠼৭ͱͳ͍ͬͯΔྖҬɼํͰಛ͕͍͠෦Ͱ ͋Δɽ Τοτܗঢ়ͷΈʹணͨ͠ GEI ͕ GMD ʹൺͯྑ͍݁ Ռ͕ग़Δ͜ͱ͕༧ଌͰ͖Δɽ Ҏ্͔ΒɼGEIɼGMDɼͦΕͧΕͷॴɼॴΛ֬ೝ͢ ΔͱͱʹɼγϧΤοτʹͮ͘جา༰ಛͱಈ͖Λ໌ࣔత ʹར༻ͨ͠า༰ಛͷॏཁੑΛ֬ೝͨ͠ɽ ͕ࠩ΄ͱΜͲݟΒΕͳ͍ɽ͜ͷ݁Ռ͔ΒมԽʹ͓͍ ͯɼಈ͖෦ʹணͨ͠ GMD ͕ GEI ʹൺͯྑ͍݁Ռ ͕ग़Δ͜ͱ͕༧ଌͰ͖Δɽ ࣍ʹɼมԽʹ͍ͭͯߟ͢Δɽ͕มԽ͢Δ߹ɼ 6. ͓ΘΓʹ ຊจͰɼ༷ʑͳঢ়گԼʹ͓͚Δา༰ೝূͷੑೳධՁ ʹ͍ͭͯड़ͨɽ࠷ॳʹɼධՁʹ༻͍Δา༰ಛʹ͍ͭ લड़ͷ௨Γɼา༰ͷಈ͖͕େ͖͘มԽ͢ΔͷͰਓͷܗঢ় ͯɼγϧΤοτʹͮ͘جಛɼϑϨʔϜؒࠩʹͮ͘جಛ ಛʹண͢Δ͜ͱ͕ॏཁͰ͋Δɽਤ 9(a)ɼ9(d) ಉҰਓ ɼΦϓςΟΧϧϑϩʔʹͮ͘جಛͷࡾछྨʹ͚આ໌ ʹΑΔҟͳΔͷาߦը૾Ͱ͋Δɽ͜ΕΒͷาߦγʔ ͨ͠ɽ࣍ʹɼর߹ख๏ͱͯ͠ɼϢʔΫϦουͮ͘جʹڑ ⓒ 2013 Information Processing Society of Japan 7 Vol.2013-CVIM-187 No.10 2013/5/30 ใॲཧֶձڀݚใࠂ IPSJ SIG Technical Report র߹ख๏ɼਖ਼४ผੳΛ༻͍ͨর߹ख๏ʹ͍ͭͯઆ໌͠ ͨɽ࠷ʹޙɼThe OU-ISIR Gait Database, The Treadmill [12] Dataset ʹΑΔ࣮ݧΛߦ͍ɼ͕มԽ͢Δঢ়گɼ͕ มԽ͢Δঢ় ͍͓ͯʹگ1 ର 1 ೝূɼ1 ର N ೝূͷੑೳධՁ Λߦͬͨɽ݁Ռͱͯ͠ɼྫ͑ GEnIɼMGEI GMD ͕ ͕มԽ͢Δঢ়͍͓ͯʹگਫ਼্͕͢Δɼ֤ঢ়ʹگ [13] ͓͍ͯదͨ͠า༰ಛΛબ͢Δ͜ͱ͕ɼา༰ೝূʹ͓͚ Δਫ਼্ʹͭͳ͕Δ͜ͱ͕͔ͬͨɽ·ͨɼ୯७ʹಛ [14] ؒͷϢʔΫϦου͔ڑΒ૬ҧΛٻΊΔख๏ʹൺͯɼ ਖ਼४ఆੳΛ༻͍֤ͯಛͷڑΛٻΊΔख๏͕ɼา༰ ೝূͷর߹ख๏ͱͯ͠ద͍ͯ͠Δ͜ͱ͕͔ͬͨɽ [15] ࠓޙͷ՝ͱͯ͠ɼ(1) େنσʔλϕʔεʹର͢Δಈ ͖ʹͮ͘جา༰ಛ CHLACɼGMD ͷੑೳධՁɼ(2) ֎ Ͱͷา༰σʔλϕʔε (ྫ͑ɼHumanID Gait Database [16] ͳͲ) ʹର͢ΔੑೳධՁɼ(3) ؍ଌําߦํ͕ม Խ͢Δঢ়ͰگͷੑೳධՁɼͷࡾ͕ͭ͛ڍΒΕΔɽ [16] Vision, pp. 257–270, October 2010. T Kobayashi and N Otsu. “Action and simultaneous multiple-person identification using cubic higher-order local auto-correlation”. Proc of the 17th International Conference on Pattern Recognition, Vol. 4, pp. 741–744, August 2004. K Bashir, T Xiang, S Gong, and Q Mary. “Gait representation using flow fields”. Proc. of the British Machine Vision Conference 2009, September 2009. H Iwama, M Okumura, Y Makihara, and Y Yagi. “The OU-ISIR Gait Database Comprising the Large Population Dataset and Performance Evaluation of Gait Recognition”. IEEE Trans. on Information Forensics and Security, Vol. 7, pp. 1511–1521, October 2012. Y. Makihara, H. Mannami, A. Tsuji, M.A. Hossain, K. Sugiura, A. Mori, and Y. Yagi. The ou-isir gait database comprising the treadmill dataset. IPSJ Trans. on Computer Vision and Applications, Vol. 4, pp. 53– 62, Apr. 2012. S Sarkar, P J Philips, Z Liu, I Robledo, P Grother, and K Bowyer. IEEE Transactions on Pattern Analysis and Machine Intelligence. ँࣙ ຊڀݚ JSPS ج൫( ڀݚS)21220003 ͷॿΛड ͚ͨͷͰ͋Δɽ ࢀߟจݙ [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] Uidai. http://uidai.gov.in/. How biometrics could change security. http: //news.bbc.co.uk/2/hi/programmes/click_ online/7702065.stm, Oct. 2008. R. Urtasun and P. Fua. “3D Tracking for Gait Characterization and Recognition”. In Automatic Face and Gesture Recognition, 2004. Proceedings. Sixth IEEE International Conference on, Vol. 1, pp. 17–22, May. 2004. S Nicholas and J Carter. “Towards pose invariant gait reconstruction”. Proc. of the IEEE International Conference on Image Processing 2005, Vol. 3, pp. 261–264, September. 2005. J Han and B Bhanu. “Individual recognition using gait energy image”. IEEE Trans. on Pattern Analysis and Machine Intelligence, pp. 316–322, February 2006. Y Makihara, R Sagawa, Y Mukaigawa, T Echigo, and Y Yagi. “Gait recognition using a view transformation model in the frequency domain”. Proc. of the 9th European Conf. on Computer Vision, pp. 151–163, May 2006. Y Makihara and Y Yagi. “Silhouette Extraction based on Iterative Spatio-temporal Local Color Transformation and Graph-Cut Segmentation”. Proc. of the 19th Int. Conf. on Pattern Recognition, pp. 1-4, Dec 2008. K Bashir, T Xiang, and S Gong. “Gait recognition using gait entropy image”. in Proc. 3rd Int. Conf. Imaging for Crime Detection and Prevention,, pp. 1–6, December 2009. Toby HW Lam, K. H. Cheung, and James NK Liu. “Gait flow image: A silhouette-based gait representation for human identification”. Pattern recognition, Vol. 44, No. 4, pp. 973–987, April 2011. K Bashir, T Xiang, and S Gong. “Gait recognition without subject cooperation”. Pattern Recognit. Letters, Vol. 31, No. 13, pp. 2052–2060, October 2010. C Wang, J Zhang, J Pu, X Yuan, and L Wang. “Chronogait image: a novel temporal template for gait recognition”. Proc. of the 11th European Conf. on Computer ⓒ 2013 Information Processing Society of Japan 8