【CanMV K230 AI视觉】人脸识别

动态测试效果可以去下面网站自己看。）

B站视频链接：已做成合集
抖音链接：已做成合集

人脸识别

前面学习过的人脸检测，只检测出人脸，而本实验要做的人脸识别，会学习人脸特征，然后比对，实现区分不同人脸。比如常用的人脸考勤机就是这样的原理。
在这里插入图片描述

先运行人脸注册代码，人脸注册代码会将CanMV U盘/

sdcard/app/tests/utils/db_img/

目录下的人脸图片进行识别，可以看到里面预存了2张人脸图片，我呢放了3张。
在这里插入图片描述

'''
实验名称：人脸注册
实验平台：01Studio CanMV K230
教程：wiki.01studio.cc
'''from libs.PipeLine import PipeLine, ScopedTiming
from libs.AIBase import AIBase
from libs.AI2D import Ai2d
import os
import ujson
from media.media import*from time import*import nncase_runtime as nn
import ulab.numpy as np
import time
import image
import aidemo
import random
import gc
import sys
import math
# 自定义人脸检测任务类classFaceDetApp(AIBase):def__init__(self,kmodel_path,model_input_size,anchors,confidence_threshold=0.25,nms_threshold=0.3,rgb888p_size=[1280,720],display_size=[1920,1080],debug_mode=0):super().__init__(kmodel_path,model_input_size,rgb888p_size,debug_mode)# kmodel路径
        self.kmodel_path=kmodel_path
        # 检测模型输入分辨率
        self.model_input_size=model_input_size
        # 置信度阈值
        self.confidence_threshold=confidence_threshold
        # nms阈值
        self.nms_threshold=nms_threshold
        self.anchors=anchors
        # sensor给到AI的图像分辨率，宽16字节对齐
        self.rgb888p_size=[ALIGN_UP(rgb888p_size[0],16),rgb888p_size[1]]# 视频输出VO分辨率，宽16字节对齐
        self.display_size=[ALIGN_UP(display_size[0],16),display_size[1]]# debug模式
        self.debug_mode=debug_mode
        # 实例化Ai2d，用于实现模型预处理
        self.ai2d=Ai2d(debug_mode)# 设置Ai2d的输入输出格式和类型
        self.ai2d.set_ai2d_dtype(nn.ai2d_format.NCHW_FMT,nn.ai2d_format.NCHW_FMT,np.uint8, np.uint8)
        self.image_size=[]# 配置预处理操作，这里使用了pad和resize，Ai2d支持crop/shift/pad/resize/affine，具体代码请打开/sdcard/app/libs/AI2D.py查看defconfig_preprocess(self,input_image_size=None):with ScopedTiming("set preprocess config",self.debug_mode >0):# 初始化ai2d预处理配置，默认为sensor给到AI的尺寸，可以通过设置input_image_size自行修改输入尺寸
            ai2d_input_size=input_image_size if input_image_size else self.rgb888p_size
            self.image_size=[input_image_size[1],input_image_size[0]]# 计算padding参数，并设置padding预处理
            self.ai2d.pad(self.get_pad_param(ai2d_input_size),0,[104,117,123])# 设置resize预处理
            self.ai2d.resize(nn.interp_method.tf_bilinear, nn.interp_mode.half_pixel)# 构建预处理流程,参数为预处理输入tensor的shape和预处理输出的tensor的shape
            self.ai2d.build([1,3,ai2d_input_size[1],ai2d_input_size[0]],[1,3,self.model_input_size[1],self.model_input_size[0]])# 自定义后处理，results是模型输出的array列表，这里使用了aidemo库的face_det_post_process接口defpostprocess(self,results):with ScopedTiming("postprocess",self.debug_mode >0):
            res = aidemo.face_det_post_process(self.confidence_threshold,self.nms_threshold,self.model_input_size[0],self.anchors,self.image_size,results)iflen(res)==0:return res
            else:return res[0],res[1]defget_pad_param(self,image_input_size):
        dst_w = self.model_input_size[0]
        dst_h = self.model_input_size[1]# 计算最小的缩放比例，等比例缩放
        ratio_w = dst_w / image_input_size[0]
        ratio_h = dst_h / image_input_size[1]if ratio_w < ratio_h:
            ratio = ratio_w
        else:
            ratio = ratio_h
        new_w =(int)(ratio * image_input_size[0])
        new_h =(int)(ratio * image_input_size[1])
        dw =(dst_w - new_w)/2
        dh =(dst_h - new_h)/2
        top =(int)(round(0))
        bottom =(int)(round(dh *2+0.1))
        left =(int)(round(0))
        right =(int)(round(dw *2-0.1))return[0,0,0,0,top, bottom, left, right]# 自定义人脸注册任务类classFaceRegistrationApp(AIBase):def__init__(self,kmodel_path,model_input_size,rgb888p_size=[1920,1080],display_size=[1920,1080],debug_mode=0):super().__init__(kmodel_path,model_input_size,rgb888p_size,debug_mode)# kmodel路径
        self.kmodel_path=kmodel_path
        # 人脸注册模型输入分辨率
        self.model_input_size=model_input_size
        # sensor给到AI的图像分辨率，宽16字节对齐
        self.rgb888p_size=[ALIGN_UP(rgb888p_size[0],16),rgb888p_size[1]]# 视频输出VO分辨率，宽16字节对齐
        self.display_size=[ALIGN_UP(display_size[0],16),display_size[1]]# debug模式
        self.debug_mode=debug_mode
        # 标准5官
        self.umeyama_args_112 =[38.2946,51.6963,73.5318,51.5014,56.0252,71.7366,41.5493,92.3655,70.7299,92.2041]
        self.ai2d=Ai2d(debug_mode)
        self.ai2d.set_ai2d_dtype(nn.ai2d_format.NCHW_FMT,nn.ai2d_format.NCHW_FMT,np.uint8, np.uint8)# 配置预处理操作，这里使用了affine，Ai2d支持crop/shift/pad/resize/affine，具体代码请打开/sdcard/app/libs/AI2D.py查看defconfig_preprocess(self,landm,input_image_size=None):with ScopedTiming("set preprocess config",self.debug_mode >0):
            ai2d_input_size=input_image_size if input_image_size else self.rgb888p_size
            # 计算affine矩阵，并设置仿射变换预处理
            affine_matrix = self.get_affine_matrix(landm)
            self.ai2d.affine(nn.interp_method.cv2_bilinear,0,0,127,1,affine_matrix)# 构建预处理流程,参数为预处理输入tensor的shape和预处理输出的tensor的shape
            self.ai2d.build([1,3,ai2d_input_size[1],ai2d_input_size[0]],[1,3,self.model_input_size[1],self.model_input_size[0]])# 自定义后处理defpostprocess(self,results):with ScopedTiming("postprocess",self.debug_mode >0):return results[0][0]defsvd22(self,a):# svd
        s =[0.0,0.0]
        u =[0.0,0.0,0.0,0.0]
        v =[0.0,0.0,0.0,0.0]
        s[0]=(math.sqrt((a[0]- a[3])**2+(a[1]+ a[2])**2)+ math.sqrt((a[0]+ a[3])**2+(a[1]- a[2])**2))/2
        s[1]=abs(s[0]- math.sqrt((a[0]- a[3])**2+(a[1]+ a[2])**2))
        v[2]= math.sin((math.atan2(2*(a[0]* a[1]+ a[2]* a[3]), a[0]**2- a[1]**2+ a[2]**2- a[3]**2))/2)if \
        s[0]> s[1]else0
        v[0]= math.sqrt(1- v[2]**2)
        v[1]=-v[2]
        v[3]= v[0]
        u[0]=-(a[0]* v[0]+ a[1]* v[2])/ s[0]if s[0]!=0else1
        u[2]=-(a[2]* v[0]+ a[3]* v[2])/ s[0]if s[0]!=0else0
        u[1]=(a[0]* v[1]+ a[1]* v[3])/ s[1]if s[1]!=0else-u[2]
        u[3]=(a[2]* v[1]+ a[3]* v[3])/ s[1]if s[1]!=0else u[0]
        v[0]=-v[0]
        v[2]=-v[2]return u, s, v
    defimage_umeyama_112(self,src):# 使用Umeyama算法计算仿射变换矩阵
        SRC_NUM =5
        SRC_DIM =2
        src_mean =[0.0,0.0]
        dst_mean =[0.0,0.0]for i inrange(0,SRC_NUM *2,2):
            src_mean[0]+= src[i]
            src_mean[1]+= src[i +1]
            dst_mean[0]+= self.umeyama_args_112[i]
            dst_mean[1]+= self.umeyama_args_112[i +1]
        src_mean[0]/= SRC_NUM
        src_mean[1]/= SRC_NUM
        dst_mean[0]/= SRC_NUM
        dst_mean[1]/= SRC_NUM
        src_demean =[[0.0,0.0]for _ inrange(SRC_NUM)]
        dst_demean =[[0.0,0.0]for _ inrange(SRC_NUM)]for i inrange(SRC_NUM):
            src_demean[i][0]= src[2* i]- src_mean[0]
            src_demean[i][1]= src[2* i +1]- src_mean[1]
            dst_demean[i][0]= self.umeyama_args_112[2* i]- dst_mean[0]
            dst_demean[i][1]= self.umeyama_args_112[2* i +1]- dst_mean[1]
        A =[[0.0,0.0],[0.0,0.0]]for i inrange(SRC_DIM):for k inrange(SRC_DIM):for j inrange(SRC_NUM):
                    A[i][k]+= dst_demean[j][i]* src_demean[j][k]
                A[i][k]/= SRC_NUM
        T =[[1,0,0],[0,1,0],[0,0,1]]
        U, S, V = self.svd22([A[0][0], A[0][1], A[1][0], A[1][1]])
        T[0][0]= U[0]* V[0]+ U[1]* V[2]
        T[0][1]= U[0]* V[1]+ U[1]* V[3]
        T[1][0]= U[2]* V[0]+ U[3]* V[2]
        T[1][1]= U[2]* V[1]+ U[3]* V[3]
        scale =1.0
        src_demean_mean =[0.0,0.0]
        src_demean_var =[0.0,0.0]for i inrange(SRC_NUM):
            src_demean_mean[0]+= src_demean[i][0]
            src_demean_mean[1]+= src_demean[i][1]
        src_demean_mean[0]/= SRC_NUM
        src_demean_mean[1]/= SRC_NUM
        for i inrange(SRC_NUM):
            src_demean_var[0]+=(src_demean_mean[0]- src_demean[i][0])*(src_demean_mean[0]- src_demean[i][0])
            src_demean_var[1]+=(src_demean_mean[1]- src_demean[i][1])*(src_demean_mean[1]- src_demean[i][1])
        src_demean_var[0]/= SRC_NUM
        src_demean_var[1]/= SRC_NUM
        scale =1.0/(src_demean_var[0]+ src_demean_var[1])*(S[0]+ S[1])
        T[0][2]= dst_mean[0]- scale *(T[0][0]* src_mean[0]+ T[0][1]* src_mean[1])
        T[1][2]= dst_mean[1]- scale *(T[1][0]* src_mean[0]+ T[1][1]* src_mean[1])
        T[0][0]*= scale
        T[0][1]*= scale
        T[1][0]*= scale
        T[1][1]*= scale
        return T
    defget_affine_matrix(self,sparse_points):# 获取affine变换矩阵with ScopedTiming("get_affine_matrix", self.debug_mode >1):# 使用Umeyama算法计算仿射变换矩阵
            matrix_dst = self.image_umeyama_112(sparse_points)
            matrix_dst =[matrix_dst[0][0],matrix_dst[0][1],matrix_dst[0][2],
                          matrix_dst[1][0],matrix_dst[1][1],matrix_dst[1][2]]return matrix_dst
# 人脸注册任务类classFaceRegistration:def__init__(self,face_det_kmodel,face_reg_kmodel,det_input_size,reg_input_size,database_dir,anchors,confidence_threshold=0.25,nms_threshold=0.3,rgb888p_size=[1280,720],display_size=[1920,1080],debug_mode=0):# 人脸检测模型路径
        self.face_det_kmodel=face_det_kmodel
        # 人脸注册模型路径
        self.face_reg_kmodel=face_reg_kmodel
        # 人脸检测模型输入分辨率
        self.det_input_size=det_input_size
        # 人脸注册模型输入分辨率
        self.reg_input_size=reg_input_size
        self.database_dir=database_dir
        # anchors
        self.anchors=anchors
        # 置信度阈值
        self.confidence_threshold=confidence_threshold
        # nms阈值
        self.nms_threshold=nms_threshold
        # sensor给到AI的图像分辨率，宽16字节对齐
        self.rgb888p_size=[ALIGN_UP(rgb888p_size[0],16),rgb888p_size[1]]# 视频输出VO分辨率，宽16字节对齐
        self.display_size=[ALIGN_UP(display_size[0],16),display_size[1]]# debug_mode模式
        self.debug_mode=debug_mode
        self.face_det=FaceDetApp(self.face_det_kmodel,model_input_size=self.det_input_size,anchors=self.anchors,confidence_threshold=self.confidence_threshold,nms_threshold=self.nms_threshold,debug_mode=0)
        self.face_reg=FaceRegistrationApp(self.face_reg_kmodel,model_input_size=self.reg_input_size,rgb888p_size=self.rgb888p_size)# run函数defrun(self,input_np,img_file):
        self.face_det.config_preprocess(input_image_size=[input_np.shape[3],input_np.shape[2]])
        det_boxes,landms=self.face_det.run(input_np)if det_boxes:if det_boxes.shape[0]==1:# 若是只检测到一张人脸，则将该人脸注册到数据库
                db_i_name = img_file.split('.')[0]for landm in landms:
                    self.face_reg.config_preprocess(landm,input_image_size=[input_np.shape[3],input_np.shape[2]])
                    reg_result = self.face_reg.run(input_np)withopen(self.database_dir+'{}.bin'.format(db_i_name),"wb")asfile:file.write(reg_result.tobytes())print('Success!')else:print('Only one person in a picture when you sign up')else:print('No person detected')defimage2rgb888array(self,img):#4维# 将Image转换为rgb888格式with ScopedTiming("fr_kpu_deinit",self.debug_mode >0):
            img_data_rgb888=img.to_rgb888()# hwc,rgb888
            img_hwc=img_data_rgb888.to_numpy_ref()
            shape=img_hwc.shape
            img_tmp = img_hwc.reshape((shape[0]* shape[1], shape[2]))
            img_tmp_trans = img_tmp.transpose()
            img_res=img_tmp_trans.copy()# chw,rgb888
            img_return=img_res.reshape((1,shape[2],shape[0],shape[1]))return  img_return
if __name__=="__main__":# 人脸检测模型路径
    face_det_kmodel_path="/sdcard/app/tests/kmodel/face_detection_320.kmodel"# 人脸注册模型路径
    face_reg_kmodel_path="/sdcard/app/tests/kmodel/face_recognition.kmodel"# 其它参数
    anchors_path="/sdcard/app/tests/utils/prior_data_320.bin"
    database_dir="/sdcard/app/tests/utils/db/"
    database_img_dir="/sdcard/app/tests/utils/db_img/"
    face_det_input_size=[320,320]
    face_reg_input_size=[112,112]
    confidence_threshold=0.5
    nms_threshold=0.2
    anchor_len=4200
    det_dim=4
    anchors = np.fromfile(anchors_path, dtype=np.float)
    anchors = anchors.reshape((anchor_len,det_dim))
    max_register_face =100#数据库最多人脸个数
    feature_num =128#人脸识别特征维度
    fr=FaceRegistration(face_det_kmodel_path,face_reg_kmodel_path,det_input_size=face_det_input_size,reg_input_size=face_reg_input_size,database_dir=database_dir,anchors=anchors,confidence_threshold=confidence_threshold,nms_threshold=nms_threshold)try:# 获取图像列表
        img_list = os.listdir(database_img_dir)for img_file in img_list:#本地读取一张图像
            full_img_file = database_img_dir + img_file
            print(full_img_file)
            img = image.Image(full_img_file)
            img.compress_for_ide()# 转rgb888的chw格式
            rgb888p_img_ndarry = fr.image2rgb888array(img)# 人脸注册
            fr.run(rgb888p_img_ndarry,img_file)
            gc.collect()except Exception as e:
        sys.print_exception(e)finally:
        fr.face_det.deinit()
        fr.face_reg.deinit()

运行后可以看到在

/sdcard/app/tests/utils/db/

目录下出现了5个人脸数据库：
在这里插入图片描述
当然自己也可以试试非人图片能不能注册

在这里插入图片描述

Only one person in a picture when you sign up
当你注册的时候，照片上只有一个人

注册后，运行下面代码可以进行识别

'''
实验名称：人脸识别
实验平台：01Studio CanMV K230
教程：wiki.01studio.cc
说明：先执行人脸注册代码，再运行本代码
'''from libs.PipeLine import PipeLine, ScopedTiming
from libs.AIBase import AIBase
from libs.AI2D import Ai2d
import os
import ujson
from media.media import*from time import*import nncase_runtime as nn
import ulab.numpy as np
import time
import image
import aidemo
import random
import gc
import sys
import math
# 自定义人脸检测任务类classFaceDetApp(AIBase):def__init__(self,kmodel_path,model_input_size,anchors,confidence_threshold=0.25,nms_threshold=0.3,rgb888p_size=[1920,1080],display_size=[1920,1080],debug_mode=0):super().__init__(kmodel_path,model_input_size,rgb888p_size,debug_mode)# kmodel路径
        self.kmodel_path=kmodel_path
        # 检测模型输入分辨率
        self.model_input_size=model_input_size
        # 置信度阈值
        self.confidence_threshold=confidence_threshold
        # nms阈值
        self.nms_threshold=nms_threshold
        self.anchors=anchors
        # sensor给到AI的图像分辨率，宽16字节对齐
        self.rgb888p_size=[ALIGN_UP(rgb888p_size[0],16),rgb888p_size[1]]# 视频输出VO分辨率，宽16字节对齐
        self.display_size=[ALIGN_UP(display_size[0],16),display_size[1]]# debug模式
        self.debug_mode=debug_mode
        # 实例化Ai2d，用于实现模型预处理
        self.ai2d=Ai2d(debug_mode)# 设置Ai2d的输入输出格式和类型
        self.ai2d.set_ai2d_dtype(nn.ai2d_format.NCHW_FMT,nn.ai2d_format.NCHW_FMT,np.uint8, np.uint8)# 配置预处理操作，这里使用了pad和resize，Ai2d支持crop/shift/pad/resize/affine，具体代码请打开/sdcard/app/libs/AI2D.py查看defconfig_preprocess(self,input_image_size=None):with ScopedTiming("set preprocess config",self.debug_mode >0):# 初始化ai2d预处理配置，默认为sensor给到AI的尺寸，可以通过设置input_image_size自行修改输入尺寸
            ai2d_input_size=input_image_size if input_image_size else self.rgb888p_size
            # 计算padding参数，并设置padding预处理
            self.ai2d.pad(self.get_pad_param(),0,[104,117,123])# 设置resize预处理
            self.ai2d.resize(nn.interp_method.tf_bilinear, nn.interp_mode.half_pixel)# 构建预处理流程,参数为预处理输入tensor的shape和预处理输出的tensor的shape
            self.ai2d.build([1,3,ai2d_input_size[1],ai2d_input_size[0]],[1,3,self.model_input_size[1],self.model_input_size[0]])# 自定义后处理，results是模型输出的array列表，这里使用了aidemo库的face_det_post_process接口defpostprocess(self,results):with ScopedTiming("postprocess",self.debug_mode >0):
            res = aidemo.face_det_post_process(self.confidence_threshold,self.nms_threshold,self.model_input_size[0],self.anchors,self.rgb888p_size,results)iflen(res)==0:return res,res
            else:return res[0],res[1]defget_pad_param(self):
        dst_w = self.model_input_size[0]
        dst_h = self.model_input_size[1]# 计算最小的缩放比例，等比例缩放
        ratio_w = dst_w / self.rgb888p_size[0]
        ratio_h = dst_h / self.rgb888p_size[1]if ratio_w < ratio_h:
            ratio = ratio_w
        else:
            ratio = ratio_h
        new_w =(int)(ratio * self.rgb888p_size[0])
        new_h =(int)(ratio * self.rgb888p_size[1])
        dw =(dst_w - new_w)/2
        dh =(dst_h - new_h)/2
        top =(int)(round(0))
        bottom =(int)(round(dh *2+0.1))
        left =(int)(round(0))
        right =(int)(round(dw *2-0.1))return[0,0,0,0,top, bottom, left, right]# 自定义人脸注册任务类classFaceRegistrationApp(AIBase):def__init__(self,kmodel_path,model_input_size,rgb888p_size=[1920,1080],display_size=[1920,1080],debug_mode=0):super().__init__(kmodel_path,model_input_size,rgb888p_size,debug_mode)# kmodel路径
        self.kmodel_path=kmodel_path
        # 检测模型输入分辨率
        self.model_input_size=model_input_size
        # sensor给到AI的图像分辨率，宽16字节对齐
        self.rgb888p_size=[ALIGN_UP(rgb888p_size[0],16),rgb888p_size[1]]# 视频输出VO分辨率，宽16字节对齐
        self.display_size=[ALIGN_UP(display_size[0],16),display_size[1]]# debug模式
        self.debug_mode=debug_mode
        # 标准5官
        self.umeyama_args_112 =[38.2946,51.6963,73.5318,51.5014,56.0252,71.7366,41.5493,92.3655,70.7299,92.2041]
        self.ai2d=Ai2d(debug_mode)
        self.ai2d.set_ai2d_dtype(nn.ai2d_format.NCHW_FMT,nn.ai2d_format.NCHW_FMT,np.uint8, np.uint8)# 配置预处理操作，这里使用了affine，Ai2d支持crop/shift/pad/resize/affine，具体代码请打开/sdcard/app/libs/AI2D.py查看defconfig_preprocess(self,landm,input_image_size=None):with ScopedTiming("set preprocess config",self.debug_mode >0):
            ai2d_input_size=input_image_size if input_image_size else self.rgb888p_size
            # 计算affine矩阵，并设置仿射变换预处理
            affine_matrix = self.get_affine_matrix(landm)
            self.ai2d.affine(nn.interp_method.cv2_bilinear,0,0,127,1,affine_matrix)# 构建预处理流程,参数为预处理输入tensor的shape和预处理输出的tensor的shape
            self.ai2d.build([1,3,ai2d_input_size[1],ai2d_input_size[0]],[1,3,self.model_input_size[1],self.model_input_size[0]])# 自定义后处理defpostprocess(self,results):with ScopedTiming("postprocess",self.debug_mode >0):return results[0][0]defsvd22(self,a):# svd
        s =[0.0,0.0]
        u =[0.0,0.0,0.0,0.0]
        v =[0.0,0.0,0.0,0.0]
        s[0]=(math.sqrt((a[0]- a[3])**2+(a[1]+ a[2])**2)+ math.sqrt((a[0]+ a[3])**2+(a[1]- a[2])**2))/2
        s[1]=abs(s[0]- math.sqrt((a[0]- a[3])**2+(a[1]+ a[2])**2))
        v[2]= math.sin((math.atan2(2*(a[0]* a[1]+ a[2]* a[3]), a[0]**2- a[1]**2+ a[2]**2- a[3]**2))/2)if \
        s[0]> s[1]else0
        v[0]= math.sqrt(1- v[2]**2)
        v[1]=-v[2]
        v[3]= v[0]
        u[0]=-(a[0]* v[0]+ a[1]* v[2])/ s[0]if s[0]!=0else1
        u[2]=-(a[2]* v[0]+ a[3]* v[2])/ s[0]if s[0]!=0else0
        u[1]=(a[0]* v[1]+ a[1]* v[3])/ s[1]if s[1]!=0else-u[2]
        u[3]=(a[2]* v[1]+ a[3]* v[3])/ s[1]if s[1]!=0else u[0]
        v[0]=-v[0]
        v[2]=-v[2]return u, s, v
    defimage_umeyama_112(self,src):# 使用Umeyama算法计算仿射变换矩阵
        SRC_NUM =5
        SRC_DIM =2
        src_mean =[0.0,0.0]
        dst_mean =[0.0,0.0]for i inrange(0,SRC_NUM *2,2):
            src_mean[0]+= src[i]
            src_mean[1]+= src[i +1]
            dst_mean[0]+= self.umeyama_args_112[i]
            dst_mean[1]+= self.umeyama_args_112[i +1]
        src_mean[0]/= SRC_NUM
        src_mean[1]/= SRC_NUM
        dst_mean[0]/= SRC_NUM
        dst_mean[1]/= SRC_NUM
        src_demean =[[0.0,0.0]for _ inrange(SRC_NUM)]
        dst_demean =[[0.0,0.0]for _ inrange(SRC_NUM)]for i inrange(SRC_NUM):
            src_demean[i][0]= src[2* i]- src_mean[0]
            src_demean[i][1]= src[2* i +1]- src_mean[1]
            dst_demean[i][0]= self.umeyama_args_112[2* i]- dst_mean[0]
            dst_demean[i][1]= self.umeyama_args_112[2* i +1]- dst_mean[1]
        A =[[0.0,0.0],[0.0,0.0]]for i inrange(SRC_DIM):for k inrange(SRC_DIM):for j inrange(SRC_NUM):
                    A[i][k]+= dst_demean[j][i]* src_demean[j][k]
                A[i][k]/= SRC_NUM
        T =[[1,0,0],[0,1,0],[0,0,1]]
        U, S, V = self.svd22([A[0][0], A[0][1], A[1][0], A[1][1]])
        T[0][0]= U[0]* V[0]+ U[1]* V[2]
        T[0][1]= U[0]* V[1]+ U[1]* V[3]
        T[1][0]= U[2]* V[0]+ U[3]* V[2]
        T[1][1]= U[2]* V[1]+ U[3]* V[3]
        scale =1.0
        src_demean_mean =[0.0,0.0]
        src_demean_var =[0.0,0.0]for i inrange(SRC_NUM):
            src_demean_mean[0]+= src_demean[i][0]
            src_demean_mean[1]+= src_demean[i][1]
        src_demean_mean[0]/= SRC_NUM
        src_demean_mean[1]/= SRC_NUM
        for i inrange(SRC_NUM):
            src_demean_var[0]+=(src_demean_mean[0]- src_demean[i][0])*(src_demean_mean[0]- src_demean[i][0])
            src_demean_var[1]+=(src_demean_mean[1]- src_demean[i][1])*(src_demean_mean[1]- src_demean[i][1])
        src_demean_var[0]/= SRC_NUM
        src_demean_var[1]/= SRC_NUM
        scale =1.0/(src_demean_var[0]+ src_demean_var[1])*(S[0]+ S[1])
        T[0][2]= dst_mean[0]- scale *(T[0][0]* src_mean[0]+ T[0][1]* src_mean[1])
        T[1][2]= dst_mean[1]- scale *(T[1][0]* src_mean[0]+ T[1][1]* src_mean[1])
        T[0][0]*= scale
        T[0][1]*= scale
        T[1][0]*= scale
        T[1][1]*= scale
        return T
    defget_affine_matrix(self,sparse_points):# 获取affine变换矩阵with ScopedTiming("get_affine_matrix", self.debug_mode >1):# 使用Umeyama算法计算仿射变换矩阵
            matrix_dst = self.image_umeyama_112(sparse_points)
            matrix_dst =[matrix_dst[0][0],matrix_dst[0][1],matrix_dst[0][2],
                          matrix_dst[1][0],matrix_dst[1][1],matrix_dst[1][2]]return matrix_dst
# 人脸识别任务类classFaceRecognition:def__init__(self,face_det_kmodel,face_reg_kmodel,det_input_size,reg_input_size,database_dir,anchors,confidence_threshold=0.25,nms_threshold=0.3,face_recognition_threshold=0.75,rgb888p_size=[1280,720],display_size=[1920,1080],debug_mode=0):# 人脸检测模型路径
        self.face_det_kmodel=face_det_kmodel
        # 人脸识别模型路径
        self.face_reg_kmodel=face_reg_kmodel
        # 人脸检测模型输入分辨率
        self.det_input_size=det_input_size
        # 人脸识别模型输入分辨率
        self.reg_input_size=reg_input_size
        self.database_dir=database_dir
        # anchors
        self.anchors=anchors
        # 置信度阈值
        self.confidence_threshold=confidence_threshold
        # nms阈值
        self.nms_threshold=nms_threshold
        self.face_recognition_threshold=face_recognition_threshold
        # sensor给到AI的图像分辨率，宽16字节对齐
        self.rgb888p_size=[ALIGN_UP(rgb888p_size[0],16),rgb888p_size[1]]# 视频输出VO分辨率，宽16字节对齐
        self.display_size=[ALIGN_UP(display_size[0],16),display_size[1]]# debug_mode模式
        self.debug_mode=debug_mode
        self.max_register_face =100# 数据库最多人脸个数
        self.feature_num =128# 人脸识别特征维度
        self.valid_register_face =0# 已注册人脸数
        self.db_name=[]
        self.db_data=[]
        self.face_det=FaceDetApp(self.face_det_kmodel,model_input_size=self.det_input_size,anchors=self.anchors,confidence_threshold=self.confidence_threshold,nms_threshold=self.nms_threshold,rgb888p_size=self.rgb888p_size,display_size=self.display_size,debug_mode=0)
        self.face_reg=FaceRegistrationApp(self.face_reg_kmodel,model_input_size=self.reg_input_size,rgb888p_size=self.rgb888p_size,display_size=self.display_size)
        self.face_det.config_preprocess()# 人脸数据库初始化
        self.database_init()# run函数defrun(self,input_np):# 执行人脸检测
        det_boxes,landms=self.face_det.run(input_np)
        recg_res =[]for landm in landms:# 针对每个人脸五官点，推理得到人脸特征，并计算特征在数据库中相似度
            self.face_reg.config_preprocess(landm)
            feature=self.face_reg.run(input_np)
            res = self.database_search(feature)
            recg_res.append(res)return det_boxes,recg_res
    defdatabase_init(self):# 数据初始化，构建数据库人名列表和数据库特征列表with ScopedTiming("database_init", self.debug_mode >1):
            db_file_list = os.listdir(self.database_dir)for db_file in db_file_list:ifnot db_file.endswith('.bin'):continueif self.valid_register_face >= self.max_register_face:break
                valid_index = self.valid_register_face
                full_db_file = self.database_dir + db_file
                withopen(full_db_file,'rb')as f:
                    data = f.read()
                feature = np.frombuffer(data, dtype=np.float)
                self.db_data.append(feature)
                name = db_file.split('.')[0]
                self.db_name.append(name)
                self.valid_register_face +=1defdatabase_reset(self):# 数据库清空with ScopedTiming("database_reset", self.debug_mode >1):print("database clearing...")
            self.db_name =[]
            self.db_data =[]
            self.valid_register_face =0print("database clear Done!")defdatabase_search(self,feature):# 数据库查询with ScopedTiming("database_search", self.debug_mode >1):
            v_id =-1
            v_score_max =0.0# 将当前人脸特征归一化
            feature /= np.linalg.norm(feature)# 遍历当前人脸数据库，统计最高得分for i inrange(self.valid_register_face):
                db_feature = self.db_data[i]
                db_feature /= np.linalg.norm(db_feature)# 计算数据库特征与当前人脸特征相似度
                v_score = np.dot(feature, db_feature)/2+0.5if v_score > v_score_max:
                    v_score_max = v_score
                    v_id = i
            if v_id ==-1:# 数据库中无人脸return'unknown'elif v_score_max < self.face_recognition_threshold:# 小于人脸识别阈值，未识别return'unknown'else:# 识别成功
                result ='name: {}, score:{}'.format(self.db_name[v_id],v_score_max)return result
    # 绘制识别结果defdraw_result(self,pl,dets,recg_results):
        pl.osd_img.clear()if dets:for i,det inenumerate(dets):# （1）画人脸框
                x1, y1, w, h =map(lambda x:int(round(x,0)), det[:4])
                x1 = x1 * self.display_size[0]//self.rgb888p_size[0]
                y1 = y1 * self.display_size[1]//self.rgb888p_size[1]
                w =  w * self.display_size[0]//self.rgb888p_size[0]
                h = h * self.display_size[1]//self.rgb888p_size[1]
                pl.osd_img.draw_rectangle(x1,y1, w, h, color=(255,0,0,255), thickness =4)# （2）写人脸识别结果
                recg_text = recg_results[i]
                pl.osd_img.draw_string_advanced(x1,y1,32,recg_text,color=(255,255,0,0))if __name__=="__main__":# 注意：执行人脸识别任务之前，需要先执行人脸注册任务进行人脸身份注册生成feature数据库# 显示模式，默认"hdmi",可以选择"hdmi"和"lcd"
    display_mode="lcd"if display_mode=="hdmi":
        display_size=[1920,1080]else:
        display_size=[800,480]# 人脸检测模型路径
    face_det_kmodel_path="/sdcard/app/tests/kmodel/face_detection_320.kmodel"# 人脸识别模型路径
    face_reg_kmodel_path="/sdcard/app/tests/kmodel/face_recognition.kmodel"# 其它参数
    anchors_path="/sdcard/app/tests/utils/prior_data_320.bin"
    database_dir ="/sdcard/app/tests/utils/db/"
    rgb888p_size=[1920,1080]
    face_det_input_size=[320,320]
    face_reg_input_size=[112,112]
    confidence_threshold=0.5
    nms_threshold=0.2
    anchor_len=4200
    det_dim=4
    anchors = np.fromfile(anchors_path, dtype=np.float)
    anchors = anchors.reshape((anchor_len,det_dim))
    face_recognition_threshold =0.75# 人脸识别阈值# 初始化PipeLine，只关注传给AI的图像分辨率，显示的分辨率
    pl=PipeLine(rgb888p_size=rgb888p_size,display_size=display_size,display_mode=display_mode)
    pl.create()
    fr=FaceRecognition(face_det_kmodel_path,face_reg_kmodel_path,det_input_size=face_det_input_size,reg_input_size=face_reg_input_size,database_dir=database_dir,anchors=anchors,confidence_threshold=confidence_threshold,nms_threshold=nms_threshold,face_recognition_threshold=face_recognition_threshold,rgb888p_size=rgb888p_size,display_size=display_size)
    clock = time.clock()try:whileTrue:
            os.exitpoint()
            clock.tick()
            img=pl.get_frame()# 获取当前帧
            det_boxes,recg_res=fr.run(img)# 推理当前帧print(det_boxes,recg_res)# 打印结果
            fr.draw_result(pl,det_boxes,recg_res)# 绘制推理结果
            pl.show_image()# 展示推理效果
            gc.collect()print(clock.fps())#打印帧率except Exception as e:
        sys.print_exception(e)finally:
        fr.face_det.deinit()
        fr.face_reg.deinit()
        pl.destroy()

主代码变量说明det_boxes人脸检测结果recg_res人脸识别结果
在这里插入图片描述

标签：人工智能

本文转载自: https://blog.csdn.net/weixin_45020839/article/details/142106041
版权归原作者 咸鱼桨 所有，如有侵权，请联系我们删除。

【CanMV K230 AI视觉】人脸识别

【CanMV K230 AI视觉】人脸识别

人脸识别

发表评论

“【CanMV K230 AI视觉】人脸识别”的评论:

关于作者

overfit同步小助手

相关阅读

文章导航

【CanMV K230 AI视觉】 人脸识别

【CanMV K230 AI视觉】 人脸识别

人脸识别

发表评论

“【CanMV K230 AI视觉】 人脸识别”的评论:

关于作者

overfit同步小助手

相关阅读

文章导航

【CanMV K230 AI视觉】人脸识别

【CanMV K230 AI视觉】人脸识别

“【CanMV K230 AI视觉】人脸识别”的评论: