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
157
158
159
160
161
162
163
164
// 霍夫变换（直线检测，纯 MoonBit）
///|
/// 检测到的直线
pub(all) struct HoughLine {
rho : Float // 极径（原点到直线的垂直距离）
theta : Float // 极角（弧度）
votes : Int // 累加器票数
}
///|
/// 霍夫直线变换。对二值图（非零为边缘点）检测直线。
/// theta_resolution 为角度分辨率（度数，默认 1.0），rho_resolution 为极径分辨率（像素，默认 1.0）。
/// threshold 为最小票数阈值，返回票数 >= threshold 的直线，按票数降序排列。
pub fn hough_lines(
img : @types.Image,
threshold : Int,
theta_resolution? : Float = 1.0F,
rho_resolution? : Float = 1.0F,
) -> Array[HoughLine] raise @types.LoadError {
// 输入验证
if img.width <= 0 || img.height <= 0 {
raise @types.LoadError::DecodeFailed(
"hough_lines: 图像尺寸必须大于0",
)
}
if img.channels != 1 && img.channels != 3 && img.channels != 4 {
raise @types.LoadError::DecodeFailed(
"hough_lines: 仅支持 channels=1/3/4",
)
}
let expected_size = img.width * img.height * img.channels
if img.data.length() != expected_size {
raise @types.LoadError::DecodeFailed(
"hough_lines: 数据大小不匹配，期望 " +
expected_size.to_string() +
"，实际 " +
img.data.length().to_string(),
)
}
if threshold <= 0 {
raise @types.LoadError::DecodeFailed("hough_lines: threshold必须大于0")
}
if theta_resolution <= 0.0F {
raise @types.LoadError::DecodeFailed(
"hough_lines: theta_resolution必须大于0",
)
}
if rho_resolution <= 0.0F {
raise @types.LoadError::DecodeFailed(
"hough_lines: rho_resolution必须大于0",
)
}
let w = img.width
let h = img.height
let ch = img.channels
// 角度范围 0~180 度
let n_theta = (180.0F / theta_resolution).to_int()
let diag = Float::from_int(w * w + h * h).sqrt()
let max_rho = (diag / rho_resolution).to_int() + 1
let n_rho = 2 * max_rho + 1
// 累加器（一维数组，提高缓存友好性）
let accumulator : Array[Int] = Array::make(n_rho * n_theta, 0)
// 预计算 sin/cos 表
let cos_table : Array[Float] = Array::make(n_theta, 0.0F)
let sin_table : Array[Float] = Array::make(n_theta, 0.0F)
let angle_scale = theta_resolution * 3.141592653589793F / 180.0F
for t in 0..<n_theta {
let angle = Float::from_int(t) * angle_scale
cos_table[t] = @math.cosf(angle)
sin_table[t] = @math.sinf(angle)
}
// 预计算 1/rho_resolution，用乘法替代除法
let inv_rho_res = 1.0F / rho_resolution
let rho_max_idx = n_rho - 1
// 预计算首通道 to_int（只看第一个通道判断边缘）
let data_int : Array[Int] = Array::make(w * h, 0)
for i in 0..<(w * h) {
data_int[i] = img.data[i * ch].to_int()
}
// 遍历边缘点
for y in 0..<h {
let fy = Float::from_int(y)
let row_base = y * w
for x in 0..<w {
if data_int[row_base + x] == 0 {
continue
}
let fx = Float::from_int(x)
for t in 0..<n_theta {
let rho = fx * cos_table[t] + fy * sin_table[t]
let raw_idx = (rho * inv_rho_res).to_int() + max_rho
// 内联 clamp，避免方法调用开销
let rho_idx = if raw_idx < 0 {
0
} else if raw_idx > rho_max_idx {
rho_max_idx
} else {
raw_idx
}
accumulator[rho_idx * n_theta + t] += 1
}
}
}
// 收集结果
let lines : Array[HoughLine] = []
let max_rho_f = Float::from_int(max_rho)
for r in 0..<n_rho {
let row_base = r * n_theta
for t in 0..<n_theta {
let votes = accumulator[row_base + t]
if votes >= threshold {
lines.push(HoughLine::{
rho: (Float::from_int(r) - max_rho_f) * rho_resolution,
theta: Float::from_int(t) * angle_scale,
votes,
})
}
}
}
// 按票数降序排序
lines.sort_by(fn(a, b) { b.votes - a.votes })
lines
}
///|
/// 非极大值抑制。对霍夫变换结果做 NMS，去除相近的重复直线。
pub fn hough_lines_nms(
lines : Array[HoughLine],
rho_threshold? : Float = 5.0F,
theta_threshold? : Float = 0.1F,
) -> Array[HoughLine] raise @types.LoadError {
// 参数验证
if rho_threshold <= 0.0F {
raise @types.LoadError::DecodeFailed(
"hough_lines_nms: rho_threshold必须大于0",
)
}
if theta_threshold <= 0.0F {
raise @types.LoadError::DecodeFailed(
"hough_lines_nms: theta_threshold必须大于0",
)
}
let result : Array[HoughLine] = []
let suppressed : Array[Bool] = Array::make(lines.length(), false)
for i in 0..<lines.length() {
if suppressed[i] {
continue
}
result.push(lines[i])
for j in (i + 1)..<lines.length() {
if suppressed[j] {
continue
}
let dr = (lines[i].rho - lines[j].rho).abs()
let dt = (lines[i].theta - lines[j].theta).abs()
if dr < rho_threshold && dt < theta_threshold {
suppressed[j] = true
}
}
}
result
}