Research Background
Optical Character Recognition (OCR) refers to the process of converting text within images into computer-editable text content. Numerous researchers have studied related technologies for a long time, resulting in many mature OCR technologies and products, such as Hanwang OCR, ABBYY FineReader, and Tesseract OCR. It is worth mentioning that ABBYY FineReader not only has high accuracy (including for Chinese recognition) but also preserves most of the original layout, making it a very powerful commercial OCR software.
However, among the many OCR products, except for Tesseract OCR, most are closed-source or even commercial software. We can neither embed them into our own programs nor improve them. The only open-source choice is Google’s Tesseract OCR, but its recognition performance is not particularly good, and the accuracy for Chinese recognition is relatively low, requiring further improvement.
In summary, whether for academic research or practical application, it is necessary to explore and improve OCR technology. Our team has divided the complete OCR system into four aspects: “Feature Extraction,” “Text Localization,” “Optical Recognition,” and “Language Model.” We are solving these step-by-step to ultimately complete a usable and comprehensive OCR system for printed text. This system can be initially used for image text recognition on platforms such as e-commerce and WeChat to verify the authenticity of the information provided.
Research Assumptions
In this article, we assume that the text portions of the images have the following characteristics:
1. We assume that the image fonts to be recognized are relatively standard printed fonts, such as Songti, Heiti, Kaiti, Xingshu, etc.;
2. There should be a clear contrast between the text and the background;
3. When designing the model, we assume that the image text is arranged horizontally;
4. The strokes of the characters should have a certain width and should not be too thin;
5. The color of a single character should, at most, be a gradient;
6. Generally, characters are formed by relatively dense strokes and often possess a certain degree of connectivity.
As can be seen, these characteristics are common features of typical e-commerce promotional posters, making these assumptions quite reasonable.
Analysis Workflow
Experimental Platform
The experiments in this article were completed in an environment of CentOS 7 + Python 2.7. Among them, the image processing part utilized the following extension libraries: Numpy, SciPy, Pandas, and Pillow. The Convolutional Neural Network (CNN) models utilized the following extension libraries: Keras and Theano. Specific experimental configurations will be discussed further later.
Reprinting: Please include the original address of this article: https://kexue.fm/archives/3781
For more details regarding reprinting, please refer to: Scientific Space FAQ