基于视觉-语言大模型的实木家具选料优选方法

打开文本图片集
中图分类号:S781.1 文献标志码:A
Abstract: 【Objective】Wood selection represents a crucial stage preceding the manufacturing processof solid wood furniture. Cuentlythldodoclsdtlalrstetetsel hasbeendevisedThisodeliscapableofsignifcantlyenhancingtheaccuracyofsolidwoodidentification,threbyestablisgaolid foundationforensuringthequalityandqualityrelatedaspectsofwoodenfurniture.【Method】Fourtypesofwoodfrequentlyemployed insolidwooduiture,melyishpine,archechndashwereelectedasteseachbectsasedoesace characteristicssuchslordtextueofteodinedstictsoldoodatasesweeconstructed.Theaskofsoldodfuue materialselectionwasaomplishedbyleveragigthetransferleaingapproachandtheimage-textpairpredictionandclasification capabilitiesofthevisualanguagelargemodelTofurtheraugmentherecognitionaccuracyofwoodforsolidwoodfuriture,the contrastivelanguage image pre-training (CLIP)visuallanguagelarge model was introducedandoptimizedusing Linear Probing. Thisoptimizationenabledtheeficientidentificationandclasifcationofmultiplesurfacefeaturesofthewood【Result】Atotalof 1 740 solid wood images were randomly partitioned into training and test sets at a ratio of 3:1 and then compared and analyzed with theResNet10lnetworkandFullFine-Tning.ByadoptingtheliearprobingmethodwithintheCLIPmodel,theoveralrcoition accuracyof themodel reach 98.29% ,themAP is 96.72% ,theiterative parameter size of themodel is 0.26MB ,and the output time per image is 37.48ms .Both theclassification performance and speed areoptimized.【Conclusion】The results indicated that in the task of solidwoodfuiturematerialselectionwithlimitedsampletrainng,theleaingperformaneandstabilityofteproposedmetodave beenefctivelyliatdIoldstottialfpiaoitellgntodleioipmntchielyo woodutilizationrateandtheproductionefciecyofsolidwoodfuiture.Moreover,itprovdessigfcantgudaceforpromotingthe intelligent manufacturing of solid wood furniture.
Keywords:solidwoodfurniturematerialselection;visuallanguagelarge model; linearprobing;inteligent manufacturing
家具作为人们日常生活中的必需品,随着生活水平的不断提升,也从基础功能需求转向品质与审美兼具。(剩余13110字)