网站建设汇报稿,广西南宁最新新闻事件,苏州快速建站模板,wordpress数据库创建行人识别yolov5和v7对比yolo车距源码:yolov5双目检测车辆识别(2023年单目双目python源码毕业设计)上盒岛APP#xff0c;开线上盲盒商店http://www.hedaoapp.com/yunPC/goodsDetails?pid4132 为了提高传统遗传算法(genetic algorithm, GA)IGA优化BP网络迭代时间过长以及精度偏…
行人识别yolov5和v7对比yolo车距源码:yolov5双目检测车辆识别(2023年单目双目python源码毕业设计)上盒岛APP开线上盲盒商店http://www.hedaoapp.com/yunPC/goodsDetails?pid4132 为了提高传统遗传算法(genetic algorithm, GA)IGA优化BP网络迭代时间过长以及精度偏低的缺陷设计了一种通过改进遗传算法(improved genetic algorithm, GA)IGA优化BP网络并进行完成双目视觉的定位计算。改进遗传算法来提升BP网络收敛能力并获得更强的全局寻优效果显著改善BP网络处理效率与精度最终促使相机获得更高定位精度以及运算速率。给出了IGA优化BP网络的双目视觉定位算法流程并开展了双目视觉定位实验。研究结果表明未优化坐标预测值误差均值为0.66 mm,优化坐标误差均值为0.08 mm。改进BP网络进行双目视觉定位精度达到0.12 mm,相对最初预测定位误差降低近0.01 mm。以BP网络来定位双目视觉精度均值是0.12 mm,以OpenCV定位的实际精度是0.10 mm。推断以神经网络双目视觉进行定位时满足双目视觉定位精度条件。 1.1 双目视觉定位BP网络结构
根据图1的方式通过BP网络进行双目视觉定位。共设定3层BP网络依次包括输入、输出与隐含层。通过左右相机对棋盘格角点区域进行拍摄获得图像数据再以横纵像素坐标构成输入神经元之后利用角点实际坐标参数组成BP网络输出神经元。隐含层中共包含9个神经元可以通过计算输入层神经元数量2倍再加1得到隐含层神经元数量。以wij、wki表示输入、输出层相对隐含层相对隐含层的权值。本研究总共包含1000个训练样本设定目标精度为0.000 1,同时控制迭代上限10 000
。
1.2 双目视觉定位IGA优化BP结构
重点研究了采用遗传算法来实现双目视觉定位时所需要克服的问题在此基础上设计得到更优的遗传算法来提升算法收敛能力并获得更强的全局寻优效果显著改善BP网络处理效率与精度最终促使相机获得更高定位精度以及运算速率。
1.2.1 改进GA选择算子
通过排序法来达到改进的目标将第i个个体经过排序处理后被选择概率Pnew通过式(1)进行计算
利用同名角点检测与匹配的方法从中提取得到表2中的部分像素坐标与坐标参数。以(u,v)表示像素坐标u、v依次对应水平与垂直方向。
3.1 精度测试
本次总共选择1000组通过相机拍摄得到的同名角点像素坐标及其对应的实际坐标数据组成训练集之后训练BP网络以及通过IGA进行优化处理的BP网络接着导入6组数据组成测试集再对相机实施定位。图4a是以未优化处理的BP网络计算得到的预测角点坐标与实际值可以发现此时两条曲线形成了相近的走势规律但也存在一定的差异性。图4b是通过IGA进行优化处理的BP网络角点坐标预测结果与实际值经对比发现此时两条曲线形成了良好贴合的状态可以确定精确的二维与坐标对应关系能够确保双目视觉获得更低定位误差。 为评价经过改进的神经网络实际定位性能通过改进BP网络构建二维与三维空间对应关系对神经网络完成1000次训练后再把二维图像角点数据输入系统中进行测试。按照反向方式记录二维图像角点空间坐标输出结果再跟定位块坐标开展差值比较得到表2所示的结果根据以上方式得到的差值作为最终定位精度。利用表2计算得到的改进BP网络进行双目视觉定位精度达到0.12 mm, 相对最初预测定位误差降低近0.01 mm。
3.2 速度测试
为了对神经网络双目视觉定位结果可靠度进行验证通过OpenCV完成双目视觉定位测试再跟定位结果进行比较。采用OpenCV传统双目视觉定位模式确定双目视觉参数再利用定位后的参数进行反向计算得到二维图像平面内角点三维空间坐标数据。再对比上述三维空间计算角点坐标和真实三维空间角点坐标由此确定定位精度。根据表2可知以BP网络来定位双目视觉时获得的精度均值是0.12 mm, 以OpenCV定位的实际精度是0.10 mm。由此可以推断以神经网络双目视觉进行定位时满足双目视觉定位精度条件。 # -*- coding: utf-8 -*-
import argparse
import time
from pathlib import Pathimport cv2
import torch
import torch.backends.cudnn as cudnn
from numpy import randomimport numpy as np
from PIL import Image, ImageDraw, ImageFontfrom models.experimental import attempt_load
from utils.datasets import LoadStreams, LoadImages
from utils.general import check_img_size, check_requirements, check_imshow, non_max_suppression, apply_classifier, \scale_coords, xyxy2xywh, strip_optimizer, set_logging, increment_path
from utils.plots import plot_one_box
from utils.torch_utils import select_device, load_classifier, time_synchronizedfrom stereo.dianyuntu_yolo import preprocess, undistortion, getRectifyTransform, draw_line, rectifyImage,\stereoMatchSGBM, hw3ToN3, DepthColor2Cloud, view_cloudfrom stereo import stereoconfig_040_2num 210 #207 209 210 211
def detect(save_imgFalse):num 210source, weights, view_img, save_txt, imgsz opt.source, opt.weights, opt.view_img, opt.save_txt, opt.img_sizewebcam source.isnumeric() or source.endswith(.txt) or source.lower().startswith((rtsp://, rtmp://, http://) )# Directoriessave_dir Path( increment_path(Path(opt.project) / opt.name, exist_okopt.exist_ok) ) # increment run(save_dir / labels if save_txt else save_dir).mkdir(parentsTrue, exist_okTrue) # make dir# Initializeset_logging()device select_device(opt.device)half device.type ! cpu # half precision only supported on CUDA# Load modelmodel attempt_load(weights, map_locationdevice) # load FP32 modelstride int(model.stride.max()) # model strideimgsz check_img_size(imgsz, sstride) # check img_sizeif half:model.half() # to FP16# Second-stage classifierclassify Falseif classify:modelc load_classifier(nameresnet101, n2) # initializemodelc.load_state_dict(torch.load(weights/resnet101.pt, map_locationdevice)[model]).to(device).eval()# Set Dataloadervid_path, vid_writer None, Noneif webcam:view_img check_imshow()cudnn.benchmark True # set True to speed up constant image size inferencedataset LoadStreams(source, img_sizeimgsz, stridestride)else:save_img Truedataset LoadImages(source, img_sizeimgsz, stridestride)print(img_size:)print(imgsz)# Get names and colorsnames model.module.names if hasattr(model, module) else model.namescolors [[random.randint(0, 255) for _ in range(3)] for _ in names]# Run inferenceif device.type ! cpu:model(torch.zeros(1, 3, imgsz, imgsz).to(device).type_as(next(model.parameters()))) # run oncet0 time.time()for path, img, im0s, vid_cap in dataset:img torch.from_numpy(img).to(device)img img.half() if half else img.float() # uint8 to fp16/32img / 255.0 # 0 - 255 to 0.0 - 1.0if img.ndimension() 3:img img.unsqueeze(0)# Inferencet1 time_synchronized()pred model(img, augmentopt.augment)[0]# Apply NMSpred non_max_suppression(pred, opt.conf_thres, opt.iou_thres, classesopt.classes, agnosticopt.agnostic_nms)t2 time_synchronized()# Apply Classifierif classify:pred apply_classifier(pred, modelc, img, im0s)# Process detectionsfor i, det in enumerate(pred): # detections per imageif webcam: # batch_size 1p, s, im0, frame path[i], %g: % i, im0s[i].copy(), dataset.countelse:p, s, im0, frame path, , im0s, getattr(dataset, frame, 0)p Path(p) # to Pathsave_path str(save_dir / p.name) # img.jpgtxt_path str(save_dir / labels / p.stem) ( if dataset.mode image else f_{frame}) # img.txts %gx%g % img.shape[2:] # print stringgn torch.tensor(im0.shape)[[1, 0, 1, 0]] # normalization gain whwhif len(det):# Rescale boxes from img_size to im0 sizedet[:, :4] scale_coords(img.shape[2:], det[:, :4], im0.shape).round()# Print resultsfor c in det[:, -1].unique():n (det[:, -1] c).sum() # detections per classs f{n} {names[int(c)]} {s * (n 1)} , # add to string# Write resultsfor *xyxy, conf, cls in reversed(det):if save_txt: # Write to filexywh (xyxy2xywh(torch.tensor(xyxy).view(1, 4)) / gn).view(-1).tolist() # normalized xywhprint(xywh x : %d, y : %d%(xywh[0],xywh[1]) )line (cls, *xywh, conf) if opt.save_conf else (cls, *xywh) # label formatwith open(txt_path .txt, a) as f:f.write((%g * len(line)).rstrip() % line \n)if save_img or view_img: # Add bbox to imagelabel f{names[int(cls)]} {conf:.2f} plot_one_box(xyxy, im0, labellabel, colorcolors[int(cls)], line_thickness3)##print label x,y zuobiao x (xyxy[0] xyxy[2]) / 2y (xyxy[1] xyxy[3]) / 2#print( %s is x: %d y: %d %(label,x,y) )height_0, width_0 im0.shape[0:2]if (x int(width_0/2) ):t3 time_synchronized()#################################stereo codep numstring #print(P is %d %p )# 读取数据集的图片#iml cv2.imread(./stereo/yolo/zuo/%szuo%d.bmp %(string,p) ) # 左图#imr cv2.imread(./stereo/yolo/you/%syou%d.bmp %(string,p) ) # 右图#iml cv2.imread(./stereo/yolo/zuo/%szuo%d.bmp %(string,p) ) # 左图#imr cv2.imread(./stereo/yolo/you/%syou%d.bmp %(string,p) ) # 右图#height_0, width_0 im0.shape[0:2]#print(width_0 %d % width_0)#print(height_0 %d % height_0)iml im0[0:int(height_0), 0:int(width_0/2)]imr im0[0:int(height_0), int(width_0/2):int(width_0) ]height, width iml.shape[0:2]#cv2.imshow(iml,iml)#cv2.imshow(imr,im0)#cv2.waitKey(0)#print(width %d % width)#print(height %d % height)# 读取相机内参和外参config stereoconfig_040_2.stereoCamera()# 立体校正map1x, map1y, map2x, map2y, Q getRectifyTransform(height, width, config) # 获取用于畸变校正和立体校正的映射矩阵以及用于计算像素空间坐标的重投影矩阵#print(Print Q!)#print(Q[2,3]:%.3f%Q[2,3])iml_rectified, imr_rectified rectifyImage(iml, imr, map1x, map1y, map2x, map2y)# 绘制等间距平行线检查立体校正的效果line draw_line(iml_rectified, imr_rectified)#cv2.imwrite(./yolo/%s检验%d.png %(string,p), line)# 消除畸变iml undistortion(iml, config.cam_matrix_left, config.distortion_l)imr undistortion(imr, config.cam_matrix_right, config.distortion_r)# 立体匹配iml_, imr_ preprocess(iml, imr) # 预处理一般可以削弱光照不均的影响不做也可以iml_rectified_l, imr_rectified_r rectifyImage(iml_, imr_, map1x, map1y, map2x, map2y)disp, _ stereoMatchSGBM(iml_rectified_l, imr_rectified_r, True) #cv2.imwrite(./yolo/%s视差%d.png %(string,p), disp)# 计算像素点的3D坐标左相机坐标系下points_3d cv2.reprojectImageTo3D(disp, Q) # 可以使用上文的stereo_config.py给出的参数#points_3d points_3d#print(x is :%.3f %points_3d[int(y), int(x), 0] )print(点 (%d, %d) 的三维坐标 (x:%.3fcm, y:%.3fcm, z:%.3fcm) % (int(x), int(y), points_3d[int(y), int(x), 0]/10, points_3d[int(y), int(x), 1]/10, points_3d[int(y), int(x), 2]/10) )count 0#try:while( (points_3d[int(y), int(x), 2] 0) | (points_3d[int(y), int(x), 2] 2500) ):count 1x countif( 0 points_3d[int(y), int(x), 2] 2300 ):breaky countif( 0 points_3d[int(y), int(x), 2] 2300 ):breakcount 1x - countif( 0 points_3d[int(y), int(x), 2] 2300 ):breaky - countif( 0 points_3d[int(y), int(x), 2] 2300 ):break#if(count%21):# x 1#else:# y 1text_cxy *cv2.putText(im0, text_cxy, (x, y) , cv2.FONT_ITALIC, 1.2, (0,0,255), 3)#print(count is %d %count)print(点 (%d, %d) 的三维坐标 (x:%.1fcm, y:%.1fcm, z:%.1fcm) % (int(x), int(y), points_3d[int(y), int(x), 0]/10, points_3d[int(y), int(x), 1]/10, points_3d[int(y), int(x), 2]/10) )dis ( (points_3d[int(y), int(x), 0] ** 2 points_3d[int(y), int(x), 1] ** 2 points_3d[int(y), int(x), 2] **2) ** 0.5 ) / 10print(点 (%d, %d) 的 %s 距离左摄像头的相对距离为 %0.1f cm %(x, y,label, dis) )text_x x:%.1fcm %(points_3d[int(y), int(x), 0]/10)text_y y:%.1fcm %(points_3d[int(y), int(x), 1]/10)text_z z:%.1fcm %(points_3d[int(y), int(x), 2]/10)text_dis dis:%.1fcm %discv2.rectangle(im0,(xyxy[0](xyxy[2]-xyxy[0]),xyxy[1]),(xyxy[0](xyxy[2]-xyxy[0])5220,xyxy[1]150),colors[int(cls)],-1);cv2.putText(im0, text_x, (xyxy[0](xyxy[2]-xyxy[0])5, xyxy[1]30), cv2.FONT_ITALIC, 1.2, (255,255,255), 3)cv2.putText(im0, text_y, (xyxy[0](xyxy[2]-xyxy[0])5, xyxy[1]65), cv2.FONT_ITALIC, 1.2, (255, 255, 255), 3)cv2.putText(im0, text_z, (xyxy[0](xyxy[2]-xyxy[0])5, xyxy[1]100), cv2.FONT_ITALIC, 1.2, (255, 255, 255), 3)cv2.putText(im0, text_dis, (xyxy[0](xyxy[2]-xyxy[0])5, xyxy[1]145), cv2.FONT_ITALIC, 1.2, (255, 255, 255), 3)t4 time_synchronized()print(fDone. ({t4 - t3:.3f}s))# Print time (inference NMS)print(f{s}Done. ({t2 - t1:.3f}s))# Stream resultsif view_img:cv2.imshow(str(p), im0)cv2.waitKey(1) # 1 millisecond# Save results (image with detections)if save_img:if dataset.mode image:cv2.imwrite(save_path, im0)else: # videoif vid_path ! save_path: # new videovid_path save_pathif isinstance(vid_writer, cv2.VideoWriter):vid_writer.release() # release previous video writerfourcc mp4v # output video codecfps vid_cap.get(cv2.CAP_PROP_FPS)w int(vid_cap.get(cv2.CAP_PROP_FRAME_WIDTH))h int(vid_cap.get(cv2.CAP_PROP_FRAME_HEIGHT))vid_writer cv2.VideoWriter(save_path, cv2.VideoWriter_fourcc(*fourcc), fps, (w, h))vid_writer.write(im0)if save_txt or save_img:s f\n{len(list(save_dir.glob(labels/*.txt)))} labels saved to {save_dir / labels} if save_txt else print(fResults saved to {save_dir}{s})print(fDone. ({time.time() - t0:.3f}s))if __name__ __main__:parser argparse.ArgumentParser()parser.add_argument(--weights, nargs, typestr, defaultlast_dead_fish_1000.pt, helpmodel.pt path(s))parser.add_argument(--source, typestr, default./shuangmu_dead_fish_011.mp4 , helpsource) # file/folder, 0 for webcamparser.add_argument(--img-size, typeint, default640, helpinference size (pixels))parser.add_argument(--conf-thres, typefloat, default0.25, helpobject confidence threshold)parser.add_argument(--iou-thres, typefloat, default0.45, helpIOU threshold for NMS)parser.add_argument(--device, default, helpcuda device, i.e. 0 or 0,1,2,3 or cpu)parser.add_argument(--view-img, actionstore_true, helpdisplay results)parser.add_argument(--save-txt, actionstore_true, helpsave results to *.txt)parser.add_argument(--save-conf, actionstore_true, helpsave confidences in --save-txt labels)parser.add_argument(--classes, nargs, typeint, helpfilter by class: --class 0, or --class 0 2 3)parser.add_argument(--agnostic-nms, actionstore_true, helpclass-agnostic NMS)parser.add_argument(--augment, actionstore_true, helpaugmented inference)parser.add_argument(--update, actionstore_true, helpupdate all models)parser.add_argument(--project, defaultruns/detect, helpsave results to project/name)parser.add_argument(--name, defaultexp, helpsave results to project/name)parser.add_argument(--exist-ok, actionstore_true, helpexisting project/name ok, do not increment)opt parser.parse_args()print(opt)check_requirements()with torch.no_grad():if opt.update: # update all models (to fix SourceChangeWarning)for opt.weights in [yolov5s.pt, yolov5m.pt, yolov5l.pt, yolov5x.pt]:detect()strip_optimizer(opt.weights)else:detect()
