ocr 0.1.0

A minimalist OCR library for Rust — from scratch, no external engine
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156

use ocr::OcrEngine;

fn ocr(text: &str, scale: u32) -> String {
    let img = ocr::render::render_text(text, scale);
    let engine = OcrEngine::new();
    let result = engine.recognize_image(&img).unwrap();
    result.text
}

#[test]
fn hello_world() {
    assert_eq!(ocr("HELLO WORLD", 8), "HELLO WORLD");
}

#[test]
fn alphabet_uppercase() {
    assert_eq!(ocr("ABCDEFGHIJKLMNOPQRSTUVWXYZ", 8), "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
}

#[test]
fn digits() {
    assert_eq!(ocr("0123456789", 8), "0123456789");
}

#[test]
fn lowercase() {
    assert_eq!(ocr("abcdefghijklmnopqrstuvwxyz", 8), "abcdefghijklmnopqrstuvwxyz");
}

#[test]
fn multiline() {
    assert_eq!(ocr("LINE ONE\nLINE TWO\nLINE THREE", 8), "LINE ONE\nLINE TWO\nLINE THREE");
}

#[test]
fn sentence_with_spaces() {
    assert_eq!(ocr("THE QUICK BROWN FOX", 8), "THE QUICK BROWN FOX");
}

#[test]
fn mixed_alnum_with_spaces() {
    assert_eq!(ocr("ABC 123 XYZ", 8), "ABC 123 XYZ");
}

#[test]
fn punctuation() {
    assert_eq!(ocr("HI! OK? YES.", 8), "HI! OK? YES.");
}

#[test]
fn small_scale() {
    assert_eq!(ocr("AB", 4), "AB");
}

#[test]
fn large_scale() {
    assert_eq!(ocr("AB", 12), "AB");
}

#[test]
fn single_char() {
    assert_eq!(ocr("A", 8), "A");
}

#[test]
fn single_digit() {
    assert_eq!(ocr("7", 8), "7");
}

#[test]
fn all_digits_recognized() {
    for d in '0'..='9' {
        let s = d.to_string();
        assert_eq!(ocr(&s, 8), s, "failed for digit '{}'", d);
    }
}

#[test]
fn all_uppercase_recognized() {
    for c in 'A'..='Z' {
        let s = c.to_string();
        assert_eq!(ocr(&s, 8), s, "failed for '{}'", c);
    }
}

#[test]
fn all_lowercase_recognized() {
    for c in 'a'..='z' {
        let s = c.to_string();
        assert_eq!(ocr(&s, 8), s, "failed for '{}'", c);
    }
}

#[test]
fn recognize_bytes() {
    let img = ocr::render::render_text("TEST", 8);
    let mut buf = std::io::Cursor::new(Vec::new());
    img.write_to(&mut buf, image::ImageFormat::Png).unwrap();
    let buf = buf.into_inner();
    let engine = OcrEngine::new();
    let result = engine.recognize_bytes(&buf).unwrap();
    assert_eq!(result.text, "TEST");
}

#[test]
fn empty_image_returns_empty() {
    let img = image::DynamicImage::ImageLuma8(
        image::GrayImage::from_pixel(100, 100, image::Luma([255u8])),
    );
    let engine = OcrEngine::new();
    let result = engine.recognize_image(&img).unwrap();
    assert!(result.text.is_empty());
}

#[test]
fn confidence_above_threshold() {
    let img = ocr::render::render_text("HELLO", 8);
    let engine = OcrEngine::new();
    let result = engine.recognize_image(&img).unwrap();
    assert!(result.confidence > 50.0, "confidence too low: {}", result.confidence);
}

#[test]
fn words_populated() {
    let img = ocr::render::render_text("HI OK", 8);
    let engine = OcrEngine::new();
    let result = engine.recognize_image(&img).unwrap();
    assert_eq!(result.words.len(), 2);
    assert_eq!(result.words[0].text, "HI");
    assert_eq!(result.words[1].text, "OK");
}

#[test]
fn generate_and_recognize_example_images() {
    let cases = [
        ("hello_world.png", "HELLO WORLD"),
        ("alphabet.png", "ABCDEFGHIJKLMNOPQRSTUVWXYZ"),
        ("digits.png", "0123456789"),
        ("lowercase.png", "abcdefghijklmnopqrstuvwxyz"),
        ("multiline.png", "LINE ONE\nLINE TWO\nLINE THREE"),
        ("sentence.png", "THE QUICK BROWN FOX"),
        ("mixed.png", "ABC 123 XYZ"),
        ("punctuation.png", "HI! OK? YES."),
    ];

    let engine = OcrEngine::new();
    for (filename, expected) in &cases {
        let img = ocr::render::render_text(expected, 8);
        let result = engine.recognize_image(&img).unwrap();
        assert_eq!(
            result.text, *expected,
            "mismatch for {}: expected '{}', got '{}'",
            filename, expected, result.text,
        );
    }
}