Click on the blue text
Follow us to make development more interesting
The following article comes from Intel Internet of Things, author: Wu Zeng Guangdong Rongxu Intelligent Technology Co., Ltd.
With the development of technology, computer vision algorithms based on deep learning dominate the field of OCR. As a key technology to enhance the digitalization process of traditional enterprises, the compound growth rate of China's optical character recognition market size from 2020 to 2027 is expected to exceed 20%, and it is expected to be nearly 10 billion yuan in 2027. Reference: "Overview of China's Character Recognition OCR Industry in 2022: The wave of industrial digital transformation drives the continued release of OCR demand" [1]
Source: Toubao Research Institute
In industrial applications, the requirements for OCR technology are higher because:
01
Compared with office document recognition scenarios with a single background and standardized fonts, industrial characters have complex backgrounds and non-standard fonts, such as concave and convex characters, curved characters, dotted printing, highly reflective backgrounds, etc. The overall recognition rate in the industry is only 90% or even higher. Low, far less than 99.9% of practical application requirements.
02
In industrial applications, in addition to the need for character recognition, there is usually also a need for character defect detection, such as incomplete text, missing printing, dirt, etc. It is hoped that the dual tasks of character recognition and character detection can be completed efficiently in one stop.
In the field of appearance defect inspection, Guangdong Rongxu Intelligent Technology Co., Ltd., as a supplier of customized visual inspection systems and integrated inspection equipment, has been working for more than ten years, implementing more than 200 kinds of AI quality inspection equipment, and integrating AI technology and industry The application experience is integrated into the graphical AI vision development platform: KVision, which quickly implements AI quality inspection equipment through code-free and transfer learning methods.
KVision software screenshot
Based on customers’ one-stop requirements for optical character recognition and character defect detection, we created the OCR-Master of Optical Character Master based on OpenVINO™. Its main functions are:
01
Optical character recognition: Mainly through deep learning algorithms to quickly realize the recognition of various fonts, and based on the OpenVINO™ optimization algorithm, it can be quickly deployed on Intel integrated graphics cards;
 |
 |
02
Character defect detection: mainly through the combination ofunsupervised algorithm + supervised algorithm to identify 100% of defective characters, focusing on "0 missed judgments"
 |
 |
 |
 |
In complex industrial scenarios, OCR-Master has the advantages of high recognition rate, simple operation, convenient model change, and can complete customer character recognition and defect detection requirements in one stop. It stands out from many visual products and is used in various details. in sub-sectors.
appendix
[1] "Overview of China's Character Recognition OCR Industry in 2022: The wave of industrial digital transformation drives the continued release of OCR demand":
https://pdf.dfcfw.com/pdf/H3_AP202211111580181892_1.pdf?1668199358000.pdf=Rongxu Intelligent official website:
https://www.gdrosmart.com/
Author contact information:
Wu Zeng +86 135 8095 8359
OpenVINO™
--END--
你也许想了解(点击蓝字查看)⬇️➡️ 开发者实战 | 介绍OpenVINO™ 2023.1：在边缘端赋能生成式AI➡️ 基于 ChatGLM2 和 OpenVINO™ 打造中文聊天助手➡️ 基于 Llama2 和 OpenVINO™ 打造聊天机器人➡️ OpenVINO™ DevCon 2023重磅回归！英特尔以创新产品激发开发者无限潜能➡️ 5周年更新 | OpenVINO™ 2023.0，让AI部署和加速更容易➡️ OpenVINO™5周年重头戏！2023.0版本持续升级AI部署和加速性能➡️ OpenVINO™2023.0实战 | 在 LabVIEW 中部署 YOLOv8 目标检测模型➡️ 开发者实战系列资源包来啦！➡️ 以AI作画，祝她节日快乐；简单三步，OpenVINO™ 助你轻松体验AIGC
➡️ 还不知道如何用OpenVINO™作画？点击了解教程。➡️ 几行代码轻松实现对于PaddleOCR的实时推理，快来get！➡️ 使用OpenVINO 在“端—边—云”快速实现高性能人工智能推理扫描下方二维码立即体验
OpenVINO™ 工具套件 2023.1Click to read the original text and experience OpenVINO 2023.1 now
The article is so exciting, are you “reading” it?