比例宽度字体易于眼睛观看并产生良好的 OCR
我想要一些关于比例宽度字体的建议,这些字体有吸引力且易于阅读,而且也易于使用 OCR 进行处理。我很乐意将我的 OCR 结果从可接受提升到优秀,而不必将所有像样的、比例宽度的字体扔到门外。
我排除的字体包括 OCR-A(等宽且可怕)、OCR-B(相当好,但等宽)和任何基于 MICR 的字体。我不是谷歌的怪物,但花了最后一个小时寻找建议 - 这就是我最终来到这里的原因。 ;-) 如果您有想法,我很想听听。
θịзθθ
I'd like recommendations on proportional-width fonts that are attractive and easy to read, but which are also easy to process with OCR. I'd love to push my OCR results from acceptable to excellent without having to throw every decent, proportional-width font out the door.
Fonts I've ruled out include OCR-A (monospaced and horrid), OCR-B (Pretty good, but monospaced) and any MICR based font. I'm not a monster at google, but have spent the last hour looking for advice - that's how I ended up here. ;-) If you've got ideas, I'd love to hear them.
Θịзηη
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(2)
这取决于您的 OCR 软件,但您应该尝试使用“单线”字体,这种字体的粗细变化很小或没有变化。
我能想到的最易读的就是 ITC American Typewriter。
使用纯 Helvetica 会发生什么?
我发现 OCR 的最大问题是紧密间隔的字母被错误组合。你能把你的字母间距设置得比平常宽一点吗?
This is going to depend on your OCR software, but you should try a 'monoline' font, one where there's little or no variation between thick and thin.
The most readable I can think of offhand is ITC American Typewriter.
What happens with plain Helvetica?
The biggest problem I've found with OCR is when letters that are tightly spaced are erroneously combined. Can you set your letterspacing to a little wider than normal?
如果您控制等式的 OCR 方面,您应该能够向 OCR 引擎提示您正在使用的特定字体,这将显着提高引擎的识别准确性。
这将使您能够根据所需的视觉外观选择字体,而不会放弃 OCR 准确性
If you control the OCR side of the equation, you should be able to hint to the OCR engine the specific font which you are using which will dramatically improve the recognition accuracy of the engine.
This would enable you to select a font based on the desired visual appearance without giving up OCR accuracy