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
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
// 感知哈希（纯 MoonBit）：ahash / dhash / phash / hamming_distance
///|
/// 均值哈希（aHash）。将图像缩放至 8×8 灰度，与均值比较生成 64 位哈希。
pub fn ahash(img : @types.Image) -> Array[Int] raise @types.LoadError {
// 输入验证
if img.width <= 0 || img.height <= 0 {
raise @types.LoadError::DecodeFailed("ahash: 图像尺寸必须大于0")
}
if img.channels < 1 || img.channels > 4 {
raise @types.LoadError::DecodeFailed("ahash: 通道数必须在1-4之间")
}
let expected_size = img.width * img.height * img.channels
if img.data.length() != expected_size {
raise @types.LoadError::DecodeFailed(
"ahash: 数据大小不匹配，期望 " +
expected_size.to_string() +
"，实际 " +
img.data.length().to_string(),
)
}
// 缩放到 8x8 灰度
let small = resize_to_gray_8x8(img)
let mut sum = 0
for i in 0..<64 {
sum += small[i]
}
let mean = sum / 64
let hash : Array[Int] = Array::make(64, 0)
for i in 0..<64 {
hash[i] = if small[i] > mean { 1 } else { 0 }
}
hash
}
///|
/// 差值哈希（dHash）。将图像缩放至 9×8 灰度，比较水平相邻像素差。
pub fn dhash(img : @types.Image) -> Array[Int] raise @types.LoadError {
// 输入验证
if img.width <= 0 || img.height <= 0 {
raise @types.LoadError::DecodeFailed("dhash: 图像尺寸必须大于0")
}
if img.channels < 1 || img.channels > 4 {
raise @types.LoadError::DecodeFailed("dhash: 通道数必须在1-4之间")
}
let expected_size = img.width * img.height * img.channels
if img.data.length() != expected_size {
raise @types.LoadError::DecodeFailed(
"dhash: 数据大小不匹配，期望 " +
expected_size.to_string() +
"，实际 " +
img.data.length().to_string(),
)
}
let small = resize_to_gray_9x8(img)
let hash : Array[Int] = Array::make(64, 0)
for row in 0..<8 {
for col in 0..<8 {
hash[row * 8 + col] = if small[row * 9 + col] > small[row * 9 + col + 1] {
1
} else {
0
}
}
}
hash
}
///|
/// 感知哈希（pHash 简化版）。将图像缩放至 8×8 灰度，用 DCT 近似：取每行均值与全局均值比较。
/// 完整 pHash 需 2D DCT，此处用简化版本（行列均值差分），足以满足去重需求。
pub fn phash(img : @types.Image) -> Array[Int] raise @types.LoadError {
// 输入验证
if img.width <= 0 || img.height <= 0 {
raise @types.LoadError::DecodeFailed("phash: 图像尺寸必须大于0")
}
if img.channels < 1 || img.channels > 4 {
raise @types.LoadError::DecodeFailed("phash: 通道数必须在1-4之间")
}
let expected_size = img.width * img.height * img.channels
if img.data.length() != expected_size {
raise @types.LoadError::DecodeFailed(
"phash: 数据大小不匹配，期望 " +
expected_size.to_string() +
"，实际 " +
img.data.length().to_string(),
)
}
let small = resize_to_gray_8x8(img)
// 简化 DCT：计算行均值和列均值
let row_means : Array[Int] = Array::make(8, 0)
let col_means : Array[Int] = Array::make(8, 0)
for r in 0..<8 {
let mut s = 0
for c in 0..<8 {
s += small[r * 8 + c]
}
row_means[r] = s / 8
}
for c in 0..<8 {
let mut s = 0
for r in 0..<8 {
s += small[r * 8 + c]
}
col_means[c] = s / 8
}
// 全局均值
let mut total = 0
for i in 0..<8 {
total += row_means[i]
}
let global_mean = total / 8
// 生成哈希：行均值和列均值与全局均值比较
let hash : Array[Int] = Array::make(64, 0)
for i in 0..<8 {
hash[i] = if row_means[i] > global_mean { 1 } else { 0 }
}
for i in 0..<8 {
hash[8 + i] = if col_means[i] > global_mean { 1 } else { 0 }
}
// 填充剩余 48 位用像素级比较
for i in 0..<48 {
hash[16 + i] = if small[i + 16] > global_mean { 1 } else { 0 }
}
hash
}
///|
/// 汉明距离。计算两个哈希的差异位数。
pub fn hamming_distance(
h1 : Array[Int],
h2 : Array[Int],
) -> Int raise @types.LoadError {
// 参数验证
if h1.length() == 0 {
raise @types.LoadError::DecodeFailed(
"hamming_distance: h1数组不能为空",
)
}
if h2.length() == 0 {
raise @types.LoadError::DecodeFailed(
"hamming_distance: h2数组不能为空",
)
}
if h1.length() != h2.length() {
raise @types.LoadError::DecodeFailed(
"hamming_distance: h1和h2长度必须相同",
)
}
let n = h1.length()
let mut dist = 0
for i in 0..<n {
if h1[i] != h2[i] {
dist += 1
}
}
dist
}
///|
fn resize_to_gray_8x8(img : @types.Image) -> Array[Int] {
let ch = img.channels
let w = img.width
let h = img.height
let out : Array[Int] = Array::make(64, 0)
for y in 0..<8 {
for x in 0..<8 {
// 最近邻采样
let sx = x * w / 8
let sy = y * h / 8
let v = if ch == 1 {
img.data[sy * w + sx].to_int()
} else {
let r = img.data[(sy * w + sx) * ch].to_int()
let g = img.data[(sy * w + sx) * ch + 1].to_int()
let b = img.data[(sy * w + sx) * ch + 2].to_int()
(r * 299 + g * 587 + b * 114) / 1000
}
out[y * 8 + x] = v
}
}
out
}
///|
fn resize_to_gray_9x8(img : @types.Image) -> Array[Int] {
let ch = img.channels
let w = img.width
let h = img.height
let out : Array[Int] = Array::make(72, 0)
for y in 0..<8 {
for x in 0..<9 {
let sx = x * w / 9
let sy = y * h / 8
let v = if ch == 1 {
img.data[sy * w + sx].to_int()
} else {
let r = img.data[(sy * w + sx) * ch].to_int()
let g = img.data[(sy * w + sx) * ch + 1].to_int()
let b = img.data[(sy * w + sx) * ch + 2].to_int()
(r * 299 + g * 587 + b * 114) / 1000
}
out[y * 9 + x] = v
}
}
out
}