When it comes to OCR recognition, everyone is familiar with it. Whether on mobile or computer, there are many corresponding applications on the market, and I have recommended quite a few.

Among them, there are quite a few software that supports OCR recognition on the computer side, with well-known ones like Adobe Acrobat DC and Abbyy, and even QQ Screenshot has OCR recognition capabilities.

There are also small recognition software like Tianruo OCR, PandaOCR, PearOCR, eSearch, and OcrHelper.

PearOCR (click to jump) focuses on web-based OCR recognition, requiring no software installation; just upload an image to the website for recognition. It is easy to use, intelligent, completely free, and very conscientious. All data is processed locally, and no images are uploaded, making safety a major feature.

eSearch (click to jump) is an open-source tool that covers all platforms with comprehensive functionality. In addition to supporting OCR recognition, it also supports screenshot, screen recording, screenshot search, and various other functions, making it a versatile tool!

OcrHelper (click to jump) is a third-party OCR software developed based on PaddleOCR, more like a scanning recognition application that does not require screenshots, but it consumes a relatively high amount of CPU.

Today, I want to recommend another OCR recognition software “Umi-OCR”, developed by the developer “Nagisa1224” from the Wuaiai Forum. I have previously recommended an older version, and now, after more than a year, it has been updated.

The V2 version is a complete upgrade, with most of the code rewritten, providing a brand new interface and more powerful functions, supporting customizable interfaces, command line, HTTP interface, and various other calling methods.

The software release package is available as a .7z compressed file or a .7z.exe self-extracting package, which can extract files on computers without compression software installed. The software requires no installation and is suitable for Windows 7 x64 and above. After extraction, click Umi-OCR.exe to start the program and run offline without the need for a network.

Umi-OCR supports multiple languages in the interface. When you first open the software, it will automatically switch languages according to your computer’s system settings. If you need to switch languages manually, please refer to the image below: global settings → language/Language.

It comes with a highly efficient offline OCR engine, which, as long as the computer’s performance is sufficient, can be faster than online OCR services.

“Paddle Engine Plugin Version” (125MB) has good performance, fast speed, and high resource consumption, suitable for high-end machines, but not compatible with Pentium, Celeron, or Atom CPUs.

“Rapid Engine Plugin Version” (95MB) is slightly slower, with low memory usage, suitable for low-end machines, and has good compatibility.

Tab Pages

Umi-OCR v2 consists of a series of flexible and easy-to-use tab pages, similar to browser tabs. The upper left corner of the tab bar allows you to switch to window top, and the upper right corner can lock the tab to prevent accidental closure during daily use.

Screenshot OCR

Open the screenshot OCR interface and directly call the shortcut key win+alt+c to take a screenshot, then win+alt+v to paste the recognition result. You can also set your own hotkeys.

Additionally, you can set different paragraph merges, where multi-line-natural paragraphs: intelligently recognize and merge text belonging to the same paragraph, suitable for the vast majority of scenarios. You can see the recognition results in the record.

Batch OCR

Supports batch importing of local images for recognition, and the recognized content can be saved in various formats such as txt / jsonl / md / csv (Excel).

There is also “QR Code Recognition”, which can read QR codes and barcodes from screenshots, pasted images, or dragged local images.

Summary

The brand new V2 version of Umi-OCR is even better, completely free, open-source, and capable of batch offline OCR software, highly recommended for use.

Download Links

Open-source address: https://github.com/hiroi-sora/Umi-OCR/releases/latest

123 Cloud Disk: https://www.123pan.com/s/aKfrVv-EOgN3.html