# Face recognition with LBP descriptors.
# See Timo Ahonen's "Face Recognition with Local Binary Patterns".
#
# Before running the example:
# 1) Download the AT&T faces database http://www.cl.cam.ac.uk/Research/DTG/attarchive/pub/data/att_faces.zip
# 2) Exract and copy the orl_faces directory to the SD card root.
import sensor, time, image ,display
import random
import pyb
import os
import ustruct
from pyb import UART
from pyb import Pin, Timer
sensor.reset() # Initialize the camera sensor.
sensor.set_pixformat(sensor.GRAYSCALE) # or sensor.GRAYSCALE
sensor.set_framesize(sensor.QQVGA2) # or sensor.QQVGA (or others)
sensor.set_windowing((128,160))
sensor.skip_frames(10) # Let new settings take affect.
sensor.skip_frames(time = 1000) #等待5s
#sensor.set_framesize(sensor.QQVGA2) # 128x160大小的特定液晶屏。
lcd = display.SPIDisplay()
#key = Pin('P6', Pin.IN, Pin.PULL_UP)
#key_pressed = False
debounce_counter = 0
#SUB = "s1"
NUM_SUBJECTS = 2 #图像库中不同人数，一共6人
NUM_SUBJECTS_IMGS = 20 #每人有20张样本图片
RED_LED_PIN = 1
BLUE_LED_PIN = 3
# 拍摄当前人脸。
img = sensor.snapshot()
lcd.write(sensor.snapshot()) # 拍照并显示图像。
#img = image.Image("singtown/%s/1.pgm"%(SUB))
d0 = img.find_lbp((0, 0, img.width(), img.height()))
face_cascade = image.HaarCascade("frontalface", stages=25)
#d0为当前人脸的lbp特征
img = None
pmin = 999999
num=0
global dist
ID_num = 3
face_num=20
state_flag=0
with open("/sd/data.txt", "r") as f:
lines = f.readlines() # 读取所有行
for line in lines:
line = line.strip() # 去除换行符
if line:
NUM_SUBJECTS, ID_num = map(int, line.split(", ")) # 解析为整数
print("图像库中人数:%d ",NUM_SUBJECTS)
print(f"图像库中人数: {NUM_SUBJECTS}, ID: {ID_num}")
def min(pmin, a, s):
global num
if a<pmin:
pmin=a
num=s
return pmin
key_state=0
#def timer_callback(timer):
# global key_pressed, debounce_counter,key,key_state
# # 读取按键状态（按下=0，松开=1）
# key_state = key.value()
# # 消抖逻辑（检测到低电平时开始计数）
# if key_state == 0: # 按键按下
# debounce_counter += 1
# if debounce_counter >= 10: # 5ms*4=20ms消抖
# key_state = key.value()
# if key_state == 1:
# ID_num += 1
# key_pressed = True
# debounce_counter = 0
rx_buff=[0,0,0,0]
state = 0
rx_flag = 0
uart_num=0
uart = UART(3, 115200, timeout_char=200)
uart.init(115200, bits=8, parity=None, stop=1)
#串口接收函数
def Receive_Prepare(data):
global state
global rx_buff
global rx_flag
global uart_num
global ID_num
if (state==0 and data==13):
state=1
if state==1 :
rx_buff[uart_num]=data
uart_num+=1
if data==93 and rx_buff[1]==1:
print("结束")
rx_flag=1
ID_num += 1
uart_num=0
state=0
if data==93 and rx_buff[1]==2:
print("结束")
rx_flag=0
uart_num=0
state=0
#tim = Timer(4, freq=200) # 200Hz = 5ms
#tim.callback(timer_callback) # 绑定回调函数
def sending_data(cx,cy,cw):
global uart;
#frame=[0x2C,18,cx%0xff,int(cx/0xff),cy%0xff,int(cy/0xff),0x5B];
#data = bytearray(frame)
data = ustruct.pack("<bbhhhb", #格式为俩个字符俩个短整型(2字节)
0x2C, #帧头1
0x12, #帧头2
int(cx), # up sample by 4 #数据1
int(cy), # up sample by 4 #数据2
int(cw), # up sample by 4 #数据1
# int(ch), # up sample by 4 #数据2
0x5B)
uart.write(data); #必须要传入一个字节数组
while 1:
if uart.any(): # 检测是否有数据
data = uart.readchar()
Receive_Prepare(data)
# print(rx_buff)
lcd.write(sensor.snapshot())
# rx_buff[1]=2
if rx_buff[1]==1 and rx_flag == 1 :
sending_data(1,0,0)
pyb.LED(RED_LED_PIN).on()
sensor.skip_frames(time = 500) # Give the user time to get ready.等待3s，准备一下表情。
lcd.write(sensor.snapshot())
#红灯灭，蓝灯亮
#machine.LED("LED_RED").off()
#machine.LED("LED_BLUE").on()
pyb.LED(RED_LED_PIN).off()
pyb.LED(BLUE_LED_PIN).on()
#保存截取到的图片到SD卡
# print(face_num)
sensor.snapshot().save("face/s%s/%s.bmp" % (ID_num, face_num) ) # or "example.bmp" (or others)
os.sync() # 确保数据写入硬件
face_num -= 1
pyb.LED(BLUE_LED_PIN).off()
#machine.LED("LED_BLUE").off()
if face_num==0:
state_flag=0
rx_flag=0
rx_buff=[0,0,0,0]
NUM_SUBJECTS+=1
sending_data(0,ID_num,1)
with open("/sd/data.txt", "w") as f: # "a"表示追加模式
# 写入字符串格式的整数（可添加换行符分隔）
f.write(f"{NUM_SUBJECTS}, {ID_num}\n") # 格式：计数,温度
# print(f"已写入数据：{NUM_SUBJECTS}, {{ID_num}}")
# print("Done! Reset the camera to see the saved image.")
elif (rx_buff[1] == 2 and rx_flag == 0):
# print("About to start detecting faces...")
# sensor.skip_frames(time=2000) # 给用户一个时间来准备
lcd.write(sensor.snapshot())
if uart.any(): # 检测是否有数据
data = uart.readchar()
# print(data)
Receive_Prepare(data)
# print(rx_buff)
# print("Now detecting faces!")
# diff = 1 # We'll say we detected a face after 10 frames.
# while diff and rx_buff[1]==0 and uart.any() == 0:
# img = sensor.snapshot()
# lcd.write(sensor.snapshot())
# if uart.any(): # 检测是否有数据
# data = uart.readchar()
# print(data)
# Receive_Prepare(data)
# print(rx_buff)
# # Threshold是介于0.0-1.0的阈值，较低值会同时提高检出率和假阳性
# # 率。相反，较高值会同时降低检出率和假阳性率。
# # scale是一个必须大于1.0的浮点数。较高的比例因子运行更快，
# # 但其图像匹配相应较差。理想值介于1.35-1.5之间。
# # scale控制匹配比例，使您可以检测较小的脸部。
# faces = img.find_features(face_cascade, threshold=0.2, scale_factor=1.4)
# if faces:
# diff -= 1
# for r in faces:
# img.draw_rectangle(r)
pyb.LED(RED_LED_PIN).on()
lcd.write(sensor.snapshot())
sensor.skip_frames(time = 3000)
img = sensor.snapshot()
d0 = img.find_lbp((0, 0, img.width(), img.height()))
pyb.LED(RED_LED_PIN).off()
pmin = 999999
for s in range(1, NUM_SUBJECTS+1):
global dist
dist = 0
lcd.write(sensor.snapshot()) # 拍照并显示图像。
for i in range(2, NUM_SUBJECTS_IMGS+1):
img = image.Image("face/s%d/%d.bmp"%(s, i))
lcd.write(sensor.snapshot()) # 拍照并显示图像。
d1 = img.find_lbp((0, 0, img.width(), img.height()))
#d1为第s文件夹中的第i张图片的lbp特征
dist += image.match_descriptor(d0, d1)#计算d0 d1即样本图像与被检测人脸的特征差异度。
print("Average dist for subject %d: %d"%(s, dist/NUM_SUBJECTS_IMGS))
pmin = min(pmin, dist/NUM_SUBJECTS_IMGS, s)#特征差异度越小，被检测人脸与此样本更相似更匹配。
print(pmin)
rx_buff=[0,0,0,0]
sending_data(0,num,NUM_SUBJECTS)
print(num) # num为当前最匹配的人的编号。