人脸属性检测

本节例程的位置在 百度云盘资料\野火K210 AI视觉相机\1-教程文档_例程源码\例程\10-KPU\face_attribute\face_attribute.py

介绍

人脸属性检测，根据给定的人脸判断其性别、表情等等。本模型实现了人脸“性别”、“是否微笑”、“是否张嘴”、“是否戴眼镜”的分析,下图为实机演示

野火logo 野火logo

例程

 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
import sensor, image, time, lcd
from maix import KPU
import gc

lcd.init()
sensor.reset(dual_buff=True)        # Reset and initialize the sensor. It will
                                    # run automatically, call sensor.run(0) to stop
sensor.set_pixformat(sensor.RGB565) # Set pixel format to RGB565 (or GRAYSCALE)
sensor.set_framesize(sensor.QVGA)   # Set frame size to QVGA (320x240)
sensor.skip_frames(time = 1000)     # Wait for settings take effect.
# sensor.set_hmirror(0)
# sensor.set_vflip(1)
clock = time.clock()                # Create a clock object to track the FPS.

anchor = (0.1075, 0.126875, 0.126875, 0.175, 0.1465625, 0.2246875, 0.1953125, 0.25375, 0.2440625, 0.351875, 0.341875, 0.4721875, 0.5078125, 0.6696875, 0.8984375, 1.099687, 2.129062, 2.425937)
kpu = KPU()
kpu.load_kmodel("/sd/KPU/yolo_face_detect/face_detect_320x240.kmodel")
kpu.init_yolo2(anchor, anchor_num=9, img_w=320, img_h=240, net_w=320 , net_h=240 ,layer_w=10 ,layer_h=8, threshold=0.5, nms_value=0.2, classes=1)

ld5_kpu = KPU()
print("ready load model")
ld5_kpu.load_kmodel("/sd/KPU/face_attribute/ld5.kmodel")

fac_kpu = KPU()
print("ready load model")
fac_kpu.load_kmodel("/sd/KPU/face_attribute/fac.kmodel")

pos_face_attr = ["Male ", "Mouth Open ", "Smiling ", "Glasses"]
neg_face_attr = ["Female ", "Mouth Closed", "No Smile", "No Glasses"]

# standard face key point position
FACE_PIC_SIZE = 128
dst_point =[(int(38.2946 * FACE_PIC_SIZE / 112), int(51.6963 * FACE_PIC_SIZE / 112)),
            (int(73.5318 * FACE_PIC_SIZE / 112), int(51.5014 * FACE_PIC_SIZE / 112)),
            (int(56.0252 * FACE_PIC_SIZE / 112), int(71.7366 * FACE_PIC_SIZE / 112)),
            (int(41.5493 * FACE_PIC_SIZE / 112), int(92.3655 * FACE_PIC_SIZE / 112)),
            (int(70.7299 * FACE_PIC_SIZE / 112), int(92.2041 * FACE_PIC_SIZE / 112)) ]

RATIO = 0.08
def extend_box(x, y, w, h, scale):
    x1_t = x - scale*w
    x2_t = x + w + scale*w
    y1_t = y - scale*h
    y2_t = y + h + scale*h
    x1 = int(x1_t) if x1_t>1 else 1
    x2 = int(x2_t) if x2_t<320 else 319
    y1 = int(y1_t) if y1_t>1 else 1
    y2 = int(y2_t) if y2_t<240 else 239
    cut_img_w = x2-x1+1
    cut_img_h = y2-y1+1
    return x1, y1, cut_img_w, cut_img_h

while 1:
    gc.collect()
    #print("mem free:",gc.mem_free())
    clock.tick()                    # Update the FPS clock.
    img = sensor.snapshot()
    kpu.run_with_output(img)
    dect = kpu.regionlayer_yolo2()
    fps = clock.fps()
    if len(dect) > 0:
        print("dect:",dect)
        for l in dect :
            x1, y1, cut_img_w, cut_img_h = extend_box(l[0], l[1], l[2], l[3], scale=RATIO) # 扩大人脸框
            face_cut = img.cut(x1, y1, cut_img_w, cut_img_h)
            a = img.draw_rectangle(l[0],l[1],l[2],l[3], color=(0, 255, 0))
            face_cut_128 = face_cut.resize(128, 128)
            face_cut_128.pix_to_ai()
            out = ld5_kpu.run_with_output(face_cut_128, getlist=True)
            #print("out:",len(out))
            face_key_point = []
            for j in range(5):
                x = int(KPU.sigmoid(out[2 * j])*cut_img_w + x1)
                y = int(KPU.sigmoid(out[2 * j + 1])*cut_img_h + y1)
                a = img.draw_cross(x, y, size=5, color=(0, 0, 255))
                face_key_point.append((x,y))
            T = image.get_affine_transform(face_key_point, dst_point)
            a = image.warp_affine_ai(img, face_cut_128, T)
            # face_cut_128.ai_to_pix()
            # img.draw_image(face_cut_128, 0,0)
            out2 = fac_kpu.run_with_output(face_cut_128, getlist=True)
            del face_key_point
            for i in range(4):
                th = KPU.sigmoid(out2[i])
                if th >= 0.7:
                    a = img.draw_string(l[0]+l[2], l[1]+i*16, "%s" %(pos_face_attr[i]), color=(255, 0, 0), scale=1.5)
                else:
                    a = img.draw_string(l[0]+l[2], l[1]+i*16, "%s" %(neg_face_attr[i]), color=(0, 0, 255), scale=1.5)
            del (face_cut_128)
            del (face_cut)

    a = img.draw_string(0, 0, "%2.1ffps" %(fps), color=(0, 60, 255), scale=2.0)
    lcd.display(img)

kpu.deinit()
ld5_kpu.deinit()
fac_kpu.deinit()

例程解析

1
2
3
import sensor, image, time, lcd
from maix import KPU
import gc

  • 导入必要的模块：sensor, image, time, lcd, KPU, 和 gc。这些模块分别用于控制摄像头、处理图像、时间管理、LCD显示、KPU加速和垃圾回收

1
2
3
4
5
6
7
8
lcd.init()
sensor.reset(dual_buff=True)        # Reset and initialize the sensor. It will
                                    # run automatically, call sensor.run(0) to stop
sensor.set_pixformat(sensor.RGB565) # Set pixel format to RGB565 (or GRAYSCALE)
sensor.set_framesize(sensor.QVGA)   # Set frame size to QVGA (320x240)
sensor.skip_frames(time = 1000)     # Wait for settings take effect.
# sensor.set_hmirror(0)
# sensor.set_vflip(1)

  • 初始化LCD显示和摄像头，设置摄像头图像格式为RGB565，分辨率设为QVGA（320x240），跳过一些帧以等待设置生效。

1
clock = time.clock()                # Create a clock object to track the FPS.

  • 定义一个时钟对象clock来跟踪帧率（FPS）。

1
anchor = (0.1075, 0.126875, 0.126875, 0.175, 0.1465625, 0.2246875, 0.1953125, 0.25375, 0.2440625, 0.351875, 0.341875, 0.4721875, 0.5078125, 0.6696875, 0.8984375, 1.099687, 2.129062, 2.425937)

  • 定义一个锚点数组anchor，用于YOLO v2目标检测算法。

1
2
3
kpu = KPU()
kpu.load_kmodel("/sd/KPU/yolo_face_detect/face_detect_320x240.kmodel")
kpu.init_yolo2(anchor, anchor_num=9, img_w=320, img_h=240, net_w=320 , net_h=240 ,layer_w=10 ,layer_h=8, threshold=0.5, nms_value=0.2, classes=1)

  • 创建一个KPU对象kpu，并加载人脸检测模型face_detect_320x240.kmodel，然后初始化YOLO v2目标检测。

1
2
3
4
5
6
7
ld5_kpu = KPU()
print("ready load model")
ld5_kpu.load_kmodel("/sd/KPU/face_attribute/ld5.kmodel")

fac_kpu = KPU()
print("ready load model")
fac_kpu.load_kmodel("/sd/KPU/face_attribute/fac.kmodel")

  • 创建两个KPU对象ld5_kpu和fac_kpu，分别加载人脸关键点检测模型ld5.kmodel和人脸属性分析模型fac.kmodel。

1
2
pos_face_attr = ["Male ", "Mouth Open ", "Smiling ", "Glasses"]
neg_face_attr = ["Female ", "Mouth Closed", "No Smile", "No Glasses"]

  • 定义两个数组pos_face_attr和neg_face_attr，分别表示正面和负面的面部属性。

1
2
3
4
5
6
7
# standard face key point position
FACE_PIC_SIZE = 128
dst_point =[(int(38.2946 * FACE_PIC_SIZE / 112), int(51.6963 * FACE_PIC_SIZE / 112)),
            (int(73.5318 * FACE_PIC_SIZE / 112), int(51.5014 * FACE_PIC_SIZE / 112)),
            (int(56.0252 * FACE_PIC_SIZE / 112), int(71.7366 * FACE_PIC_SIZE / 112)),
            (int(41.5493 * FACE_PIC_SIZE / 112), int(92.3655 * FACE_PIC_SIZE / 112)),
            (int(70.7299 * FACE_PIC_SIZE / 112), int(92.2041 * FACE_PIC_SIZE / 112)) ]

  • 定义一个标准人脸关键点位置的数组dst_point，这些关键点用于人脸对齐。

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
RATIO = 0.08
def extend_box(x, y, w, h, scale):
    x1_t = x - scale*w
    x2_t = x + w + scale*w
    y1_t = y - scale*h
    y2_t = y + h + scale*h
    x1 = int(x1_t) if x1_t>1 else 1
    x2 = int(x2_t) if x2_t<320 else 319
    y1 = int(y1_t) if y1_t>1 else 1
    y2 = int(y2_t) if y2_t<240 else 239
    cut_img_w = x2-x1+1
    cut_img_h = y2-y1+1
    return x1, y1, cut_img_w, cut_img_h

  • 定义一个函数extend_box，用于扩展人脸框的大小

 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
while 1:
    gc.collect()
    #print("mem free:",gc.mem_free())
    clock.tick()                    # Update the FPS clock.
    img = sensor.snapshot()
    kpu.run_with_output(img)
    dect = kpu.regionlayer_yolo2()
    fps = clock.fps()
    if len(dect) > 0:
        print("dect:",dect)
        for l in dect :
            x1, y1, cut_img_w, cut_img_h = extend_box(l[0], l[1], l[2], l[3], scale=RATIO) # 扩大人脸框
            face_cut = img.cut(x1, y1, cut_img_w, cut_img_h)
            a = img.draw_rectangle(l[0],l[1],l[2],l[3], color=(0, 255, 0))
            face_cut_128 = face_cut.resize(128, 128)
            face_cut_128.pix_to_ai()
            out = ld5_kpu.run_with_output(face_cut_128, getlist=True)
            #print("out:",len(out))
            face_key_point = []
            for j in range(5):
                x = int(KPU.sigmoid(out[2 * j])*cut_img_w + x1)
                y = int(KPU.sigmoid(out[2 * j + 1])*cut_img_h + y1)
                a = img.draw_cross(x, y, size=5, color=(0, 0, 255))
                face_key_point.append((x,y))
            T = image.get_affine_transform(face_key_point, dst_point)
            a = image.warp_affine_ai(img, face_cut_128, T)
            # face_cut_128.ai_to_pix()
            # img.draw_image(face_cut_128, 0,0)
            out2 = fac_kpu.run_with_output(face_cut_128, getlist=True)
            del face_key_point
            for i in range(4):
                th = KPU.sigmoid(out2[i])
                if th >= 0.7:
                    a = img.draw_string(l[0]+l[2], l[1]+i*16, "%s" %(pos_face_attr[i]), color=(255, 0, 0), scale=1.5)
                else:
                    a = img.draw_string(l[0]+l[2], l[1]+i*16, "%s" %(neg_face_attr[i]), color=(0, 0, 255), scale=1.5)
            del (face_cut_128)
            del (face_cut)

    a = img.draw_string(0, 0, "%2.1ffps" %(fps), color=(0, 60, 255), scale=2.0)
    lcd.display(img)

  • 进入主循环，首先进行垃圾回收，然后更新FPS时钟，捕获一帧图像，运行人脸检测模型，获取检测结果。

  • 如果检测到人脸，对人脸进行框选，并裁剪出人脸图像。

  • 对裁剪出的人脸图像进行关键点检测，然后进行人脸对齐。

  • 运行人脸属性分析模型，获取分析结果。

  • 根据分析结果，将人脸属性文字显示在图像上。

  • 显示FPS信息，并将处理后的图像显示在LCD上。

1
2
3
kpu.deinit()
ld5_kpu.deinit()
fac_kpu.deinit()

  • 循环结束后，释放所有KPU资源。