[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-657141-105":59,"doc-detail-657141-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","imago-open-source-toolkit-for-2d-chemical-structure-image-recognition","Imago - open-source toolkit for 2D chemical structure image recognition","","Different chemical databases store molecular structures as images linked to scientific articles and patents, which makes maintaining consistent, machine-usable records difficult. Imago presents an open-source toolkit for 2D chemical structure image recognition, aiming to transform structure images into a diagram graph representation usable by search and related software systems. The paper details the algorithmic approach implemented in Imago and discusses possible improvements informed by participation in Image2Structure (TRECCHEM 2011).",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/imago-open-source-toolkit-for-2d-chemical-structure-image-recognition/657141/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/imago-open-source-toolkit-for-2d-chemical-structure-image-recognition/657141.png","ImageObject",300,407,{"name":92,"@type":93},"miraaa_2004","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-30","2026-10-26",true,{"@type":102,"interactionType":103,"userInteractionCount":24},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What is Imago designed to do?","Question",{"text":112,"@type":113},"Imago is an open-source toolkit that recognizes 2D chemical structure images and converts them into a graph-like diagram structure for use in software applications, including search systems.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"Why is molecule image recognition important in chemical databases?",{"text":117,"@type":113},"Chemical databases often store molecular structures as images linked to articles and patents, so keeping such data consistent is difficult; efficient and accurate recognition algorithms are needed to make the images machine-interpretable.",{"name":119,"@type":110,"acceptedAnswer":120},"What are the main steps in Imago’s basic recognition routine?",{"text":121,"@type":113},"The routine binarizes the image, segments it into connected components using SuperSegmentation and local segmentation, removes single down stereo bonds, then classifies segments into symbols (atom labels) and graphics (bonds) before grouping and further recognition.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},657141,1793102506,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":24,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":39,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":143,"read_time":46},962094249165,"https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0","Imago: open-source toolkit for 2D chemical structure  \nimage recognition  \nViktor Smolov*, Fedor Zentsev and Mikhail Rybalkin  \nGGA Software Services LLC  \nAbstract  \nDifferent chemical databases contain molecule structures with links to articles and patents, where such molecules are presented as images. Keeping such a database in a consistent state is a challenging problem, and, thus, efficient and accurate molecule image recognition algorithms are very important for solving this task.  \nWe present an open-source toolkit for 2D chemical structure image recognition, called Imago. The main aim of this paper is to describe the algorithm approach implemented in Imago, and to share our ideas of possible improvements in the algorithm after we have summarized the results from the participation in the Image2Structure task at TRECCHEM 2011 .  \n1 Introduction  \nSignificant amount of information about chemical structure properties can be found in the scientific literature. On the one hand, there are efficient modern full text search techniques that provide the ability to search for text over a huge amount of text data. For example, the biggest database, that provides a chemical-specific text search (with automatic synonyms detection) over more than 21 million citations for biomedical literature, is the PubMed database [1] . On the other hand, there are efficient methods, that are used to search for a chemical structure with required properties (that contains special query substructure) over o huge amount of molecule structures. For example, another popular database, that provides (sub)structure search functionality over more than 31 million chemical molecules, is the PubChem database [2] . But chemical articles contains both text and molecule structure images; we can only imagine what opportunities would we get by combining text data mining methods and cheminformatics search techniques.  \nThis search problem is very similar to the so called Deep Web indexing problem: indexing the web content, that is not suitable for indexing by standard search engines. For example, Deep Web includes image and video files. According to rough estimates Deep Web is much larger than the web content, indexed by search engines.  \nDevelopment of a universal chemical search engine, that could search by both text and (sub)structures, is a challenging problem. This problem involves subproblems from different scientific areas, which include natural language processing to find the meaning of a sentence describing the interaction between different structures, and optical image recognition to find  \n* Corresponding author. E-mail address: [vsmolov@ggasoftware.com](vsmolov@ggasoftware.com), [smolovv@gmail.com](smolovv@gmail.com)  \n\n| Pseudocode 1 Overall recognition procedure |\n| --- |\n| Recognize(I)\u003Cbr>1 􀀃 􀁉—depiction of a structure, 􀁍—reconstructed molecule\u003Cbr>2 􀁍 = ?\u003Cbr>3\u003Cbr>4 􀀃 Simple filtering steps (blur, binarization)\u003Cbr>5 ProcessFilters(I) 6\u003Cbr>7 􀀃 Finding the molecule-like part of the image.\u003Cbr>8 􀀃 All further steps use 􀁉􀁍 .\u003Cbr>9 􀁉􀁍 = SuperSegmentation(I) 10\u003Cbr>11 􀁓 = Segmentation(􀁉􀁍) 12\u003Cbr>13 􀀃 Locating single down stereo bonds. All corresponding segments are removed from 􀁓 .\u003Cbr>14 􀁍 ← ExtractSingleDownBonds(S) 15\u003Cbr>16 􀀃 All image segments in 􀁓 are separated in two lists: symbols (􀁌􀁳 ) and graphics (􀁌􀁧) .\u003Cbr>17 􀁌 􀁳 ,􀁌􀁧 ← Separation(S) 18\u003Cbr>19 􀀃 All symbols combine into groups (atom labels, superatoms) . No recognition is done yet.\u003Cbr>20 􀁍 ← GroupLabels(􀁌􀁳) 21\u003Cbr>22 􀀃 Detecting line segments, aromatic circles in 􀁌􀁧 . Constructing molecule skeleton.\u003Cbr>23 􀁍 ← ExtractGraph(􀁌􀁧) 24\u003Cbr>25 􀀃 Finding single-up bonds in 􀁉􀁍 and marking corresponding bonds in the molecule.\u003Cbr>26 􀁍 ← FetchSingleUpBonds(􀁉􀁍) 27\u003Cbr>28 AttachAndRecognizeLabels(M) 29\u003Cbr>30 􀀃 Repairing or figuring out stereo bonds orientation and aromatizing molecule if circles were presented.\u003Cbr>31 􀁍 ← FixStereoCenters()\u003Cbr>32 Aromatize(M) |\n\nmolecule structure by its image from an article. This paper is ","cbCairhZo9gbDIRJ","https://ap.wps.com/l/cbCairhZo9gbDIRJ","pdf",239362,"English","# Introduction\n## Related chemical search and deep web indexing\n# Overall recognition procedure\n## Filtering and segmentation\n## Stereo bond handling and symbol/graphics separation\n## Graph extraction and label recognition\n# Basic Recognition Routine\n## Binarization and segmentation\n## Stereo bond removal\n## Segment classification into symbols and graphics","[{\"question\":\"What is Imago designed to do?\",\"answer\":\"Imago is an open-source toolkit that recognizes 2D chemical structure images and converts them into a graph-like diagram structure for use in software applications, including search systems.\"},{\"question\":\"Why is molecule image recognition important in chemical databases?\",\"answer\":\"Chemical databases often store molecular structures as images linked to articles and patents, so keeping such data consistent is difficult; efficient and accurate recognition algorithms are needed to make the images machine-interpretable.\"},{\"question\":\"What are the main steps in Imago’s basic recognition routine?\",\"answer\":\"The routine binarizes the image, segments it into connected components using SuperSegmentation and local segmentation, removes single down stereo bonds, then classifies segments into symbols (atom labels) and graphics (bonds) before grouping and further recognition.\"}]","Imago - open-source toolkit for 2D chemical structure image recognition | PDF",1792974755]