duzexu / Screenotate

支持 OCR 文字识别的载屏笔记 Mac 完整应用

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Screenotate

Screenotate is a tool by Omar Rizwan for taking annotated screenshots on Mac OS X.

You take a screenshot of a region on the screen pretty much as usual, and then Screenotate automatically detects and saves the surrounding context, with no effort needed from you.

Download it in the Releases section.

  1. Take a screenshot almost exactly as you always have: press a key shortcut (Cmd-Shift-5 by default -- compare to Cmd-Shift-4, the built-in shortcut) and then drag out a region.

  2. Then Screenotate automatically figures out what app and window you're in, what Web page the content comes from (if applicable), and OCRs the text in the screenshot to make it searchable later.

    All this information is embedded in the screenshot, which is stored as an HTML file with the image as a data-URI inside.

  3. The screenshot is sent to your Desktop by default, but you can set up a Dropbox account in Preferences so that shared links are copied to the Clipboard.

You may find it useful for snipping paragraphs or phrases out of Web pages -- if your screenshots are automatically annotated, you don't need to trawl through them later and try to figure out where they come from.

Notes, warnings, and known issues

  • OCR will not work properly on screenshots from non-Retina displays. They just don't have high enough resolution.

  • Chrome and Safari will work with URL capture out of the box. Install and enable the MozRepl extension if you want Screenotate to capture URLs from Firefox. (This feature is experimental.)

  • The app only considers the window containing the top-left corner of your selected region when it looks for metadata.

  • Though Screenotate can detect the window title for any window (active or inactive), its URL detection is not as solid: if you take a screenshot from any browser window, it will capture the URL of the frontmost window of that browser.

    (So Screenotate may leak your information -- read over your screenshots before sharing to be sure nothing gets out.)

  • Your frontmost window will temporarily lose focus as you select the screenshot region, but don't worry: in the actual screenshot, it will appear focused.

  • Upload to Dropbox is slow, and the link isn't copied until the upload finishes.

  • Some windows, including Emacs', give us blank titles for some reason.

Credits and building tips

Remember to initialize the submodules. Then just build the Xcode project.

This project uses MASShortcut, OAuth2 (to talk to Dropbox's v1 API without having to use a secret key), NSBundle-LoginItem, and most important, Tesseract (for OCR).

The camera icon is from Entypo.

I built Tesseract as a static library, and it needs leptonica, libpng, and a bunch of other libraries; the process is somewhat tricky, but a normal dynamically linked build should work. Note that Screenotate talks to the Tesseract binary (which should be put in Resources along with tessdata), instead of linking to its C++ API; it's just easier to program that way.

License

Screenotate is licensed under the GNU General Public License, version 3.0.

About

支持 OCR 文字识别的载屏笔记 Mac 完整应用

License:GNU General Public License v3.0


Languages

Language:Swift 99.0%Language:Objective-C 1.0%