Abstract：Mobile visual search model for Dunhuang murals in the smart library Purpose The purpose of this paper is to build a mobile visual search service system for the protection of Dunhuang cultural heritage in the smart libraryA novel mobile visual search model for Dunhuang murals is proposed to help users acquire rich knowledge and services convenientlyDesignmethodologyapproach Firstlocal and global features of images are extractedand the visual dictionary is generated by the k-means clusteringSecondthe mobile visual search model based on the bag-of-wordsBOWand multiple semantic associations is constructedThirdthe mobile visual search service system of the smart library is designed in the cloud environmentFurthermoreDunhuang mural images are collected to verify this modelFindings The findings reveal that the BOW_SIFT_HSV_MSA model has better search performance for Dunhuang mural images when the scale-invariant feature transformSIFTand the huesaturation and valueHSVare used to extract local and global features of the imagesCompared with different methodsthis model is the most effective way to search images with the semantic association in the topictime and space dimensionsResearch limitationsimplications Dunhuang mural image set is a part of the vast resources stored in the smart libraryand the fine-grained semantic labels could be applied to meet diverse search needsOriginalityvalue The mobile visual search service system is constructed to provide users with Dunhuang cultural services in the smart libraryA novel mobile visual search model based on BOW and multiple semantic associations is proposedThis study can also provide references for the protection and utilization of other cultural heritagesRetrieval and IndexingThe mobile phone visual search service system in the study is designed for the protection of the Dunhuang cultural heritage in the smart librarywhich helps users search for Dunhuang muralsand provide it with visual content and multi-semantic associationMore effective search resultsPattern RecognitionThe focus of this research is to establish a mobile phone visual search service system for Dunhuang cultural heritage protectionBy extracting the local and global characteristics of the imageand using the K-MEANS cluster to generate a visual dictionaryand then based on the mobile visual search model established by the word packageBOWand the multi-semantic associationit provides users with convenient acquisition of rich knowledge and servicesmethod