『您的瀏覽器不支援JavaScript功能,若網頁功能無法正常使用時,請開啟瀏覽器JavaScript狀態』

跳到主要內容區塊

工業技術研究院

:::

技術名稱: AI文件辨識

技術簡介

本所以深度學習技術為核心,開發出AI文件辨識系統,可針對特定表格化文件做自動化辨識處理:一次拍照欲辨識之多樣文件(同一個A4文件/照片上有多項文件如身份證、汽機車駕照、汽機車行照、護照、存摺、登機證、發票...等),本AI文件辨識系統可直接自動完成文件切割及分類,並調整圖像角度,切出欄位後進行OCR及資料後處理,進而將辨識影像資料結構化,方便企業整合與應用,進行自動化文件辨識作業,減少人員處理時間。

Abstract

AI DNN is to develop an Intelligent Document Processing system. It can perform automatic identification processing document. There are many documents such as passbooks, boarding passes, invoices, etc. This AI document recognition system can automatically complete document cutting and classification directly, and adjust the angle of the image and cut out the fields in the image and OCR, and then structure the recognition image data to facilitate enterprises to integrate applications for automated document recognition operations and reduce personnel processing time.

技術規格

1. 此模型可輸入PDF/Image文件檔案,輸出EXCEL或客製化處理,模型處理速度依使用之應用系統、模型複雜度與硬體效能而不同。 2. 如非本所已俱備之文件辨識種類模型,需另行討論並需俱備可訓練文件量至少數千至萬張。

Technical Specification

1. This model can be imported into PDF / Image file and exported to EXCEL or customized processing. The processing speed of the model depends on the application system used, model complexity and hardware performance. 2. If it is not the kind of document identification model already provided by ITRI, it needs to be discussed separately and a trainable document volume of at least thousands to tens of thousands.

技術特色

因應智慧手機拍照的普遍,民眾拍照上傳文件已成趨勢,因拍照方式、背景五花八門,傳統OCR技術已無法滿足,工研院因應此需求開發IDP(AI DNN)技術,解決市場上文件辨識的痛點與缺口。工研院IDP技術可處理在同一張照片裡有多張文件/證件,即使各文件/證件位置不同、拍照角度不同、文件背景五花八門、各家文件格式不同、欄位位置不同皆可辨識;可一次盲掃欲辨識之多樣文件,可直接自動完成文件切割、分類、拉正、欄位切出與辨識等。透過大量資料的訓練,能大大提昇其辨識率,相較於傳統OCR是做不到的,目前已開發多項特定文件辨識模型。可依客戶需求進行特定文件之模型訓練及系統建置。相對市面OCR 解決方案有更高的文字辨識率。

應用範圍

金融單位之文件辨識、證件辨識、KYC文件: 1. 銀行之存摺、扣繳憑單、所得清單。 2. 人壽之醫療單據、各式申請單。 3. 產險之登機證、存摺、護照、車險理賠等各式單據、車廠保養單據等。 4. 各式證件包括身份證件、汽機車駕照、汽機車行照等。

接受技術者具備基礎建議(設備)

*軟體:Nvidia GPU運行需用套件 *硬體:Nvidia GPU Server - 推論伺服器硬體規格建議: 1. CPU: Intel® Xeon® Silver 4216 處理器; 22M 快取記憶體; 2.10 GHz; x86架構同等或更高規格等級 2. RAM: DDR4-2400 ECC (512GB以上) 3. SSD INTEL® SSD D3-S4510 系列; INTEL® SSD D3-S4610 系列 (960GB以上) 4. HDD 高容量企業級HDD (視需求添購) 5. LAN 10GBase-T LAN Port

接受技術者具備基礎建議(專業)

1. AI DNN technology background. 2. AI ML/DL technology background. 3. CV、OCR、NLP background. 4. Python/C Programming.

技術分類 資訊

聯絡資訊

聯絡人:徐福裕 技術推廣組

電話:+886-3-5914566 或 Email:ben.hsu@itri.org.tw

客服專線:+886-800-45-8899

傳真:+886-3-5820240

[{"text":"企業網","weight":13.0},{"text":"材化所","weight":11.5},{"text":"機械所","weight":10.0},{"text":"綠能所","weight":9.4},{"text":"生醫所","weight":8.0},{"text":"半導體","weight":6.2},{"text":"南分院","weight":5.0},{"text":"太陽能","weight":5.0},{"text":"課程","weight":5.0},{"text":"遠紅外線","weight":5.0},{"text":"雷射","weight":4.0},{"text":"LED","weight":4.0},{"text":"LED可見光","weight":3.0},{"text":"5G","weight":3.0},{"text":"工研人","weight":3.0},{"text":"電光所","weight":3.0},{"text":"綠能與環境研究所","weight":3.0},{"text":"機械","weight":3.0},{"text":"資通所","weight":2.0},{"text":"面板","weight":2.0},{"text":"文字轉語音","weight":2.0},{"text":"實習","weight":2.0},{"text":"無人機","weight":2.0},{"text":"生醫","weight":2.0},{"text":"3D","weight":2.0},{"text":"v2x","weight":2.0},{"text":"員工","weight":2.0},{"text":"地圖","weight":2.0},{"text":"太陽光電","weight":2.0},{"text":"材料與化工研究所","weight":1.0}]