Question

我使用tesseract的C＃包装器 - charlesw。并且无法理解，为什么字典验证不起作用。如何启用字典验证？例如＆＃34; EXOMPLE＆＃34; - 不是词典单词，更改＆＃34;示例＆＃34;

            TesseractEngine engine = new TesseractEngine(@"tessdata/", "rus", EngineMode.TesseractOnly);

        engine.SetVariable("tessedit_char_whitelist", whiteList);

        // Tesseract Settings    
        engine.SetVariable("language_model_penalty_non_dict_word", 0.65);
        engine.SetVariable("language_model_penalty_non_freq_dict_word", 0.6);

Answer 1

Charles在构造函数中添加“config”参数。现在我们可以使用字典验证

TesseractEngine engine = new TesseractEngine(@"tessdata/", "rus", EngineMode.TesseractOnly, "config");

配置：

load_system_dawg F
load_freq_dawg F
user_words_suffix user-words
user_patterns_suffix user-patterns

如何在tesseract中启用字典验证

1 个答案: