windfly_al的博客

vim 进化版使用于windows，版本 8.0.281.0 ,亲测可用～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～

Caffe2 优秀的开元图像处理架构 第三方包，包括： benchmark-4bf28e611b55de8a2d4eece3c335e014f8b0f630.zip cnmem-28a182d49529da49f4ac4e3941cec3edf16b3540.zip cub-01347a797c620618d09e7d2d90bce4be4c42513e.zip eigen-ae9889a130bd0a9d3007f41d015563c2e8ac605f.zip FP16-2e9eeeb0b463736d13b887d790ac7e72e78fa4bc.zip FXdiv-8f85044fb41e560508cd69ed26c9afb9cc120e8a.zip gloo-530878247b04c423fd35477208f68e70b8126e2d.zip googletest-5e7fd50e17b6edf1cadff973d0ec68966cf3265e.zip ios-cmake-e24081928d9ceec4f4adfcf12293f1e2a20eaedc.zip nccl-2a974f5ca2aa12b178046b2206b43f1fd69d9fae.zip nervanagpu-d4eefd50fbd7d34a17dddbc829888835d67b5f4a.zip NNPACK-02bfa475d64040cd72b7c01daa9e862523ae87da.zip protobuf-a428e42072765993ff674fda72863c9f1aa2d268.zip psimd-0b26a3fb98dd6af7e1f4e0796c56df6b32b1c016.zip pthreadpool-9e17903a3fc963fe86b151aaddae7cf1b1d34815.zip pybind11-f38f359f96815421f1780c1a676715efd041f1ae.zip xware-desktop_0.13.20160328_amd64.deb ～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～～`

( ffmpeg-3.3.3.tar.bz2 解压后编译安装) 1.下载 ffmpeg-*.tar.gz 到 Ffmpeg 官网 https://ffmpeg.org/download.html 挑选你要升级到的版本，然后下载，比如作者下载的是 ffmpeg-2.0.tar.gz。 2.编译安装 (注： 如果执行./configure报yasm错 则下载 yasm-1.3.0.tar.gz 执行 tar zxvf yasm-1.3.0.tar.gz cd yasm-1.3.0 ./configure make && make install) tar -zxvf ffmpeg-2.0.tar.gz cd ffmpeg-2.0 ./configure --enable-shared --prefix=/usr/local/ffmpeg make make install 3.动态链接库 vi /etc/ld.so.conf 加入：/usr/local/ffmpeg/lib 执行 ldconfig 4.为 Ffmpeg 加入环境变量 vi /etc/profile 加入以下内容: FFMPEG=/usr/local/ffmpeg PATH加入:$FFMPEG/bin 5.使修改立即生效 source /etc/profile 执行 ffmpeg -version 打印结果 ffmpeg version 2.0 built on Jul 24 2013 09:59:06 with gcc 4.4.7 (GCC) 20120313 (Red Hat 4.4.7-3) configuration: --enable-shared --prefix=/usr/local/ffmpeg libavutil 52. 38.100 / 52. 38.100 libavcodec 55. 18.102 / 55. 18.102 libavformat 55. 12.100 / 55. 12.100 libavdevice 55. 3.100 / 55. 3.100 libavfilter 3. 79.101 / 3. 79.101 libswscale 2. 3.100 / 2. 3.100 libswresample 0. 17.102 / 0. 17.102 证明已升级成功。如果遇到 ffmpeg: error while loading shared libraries: libavdevice.so.55: cannot open shared object file: No such file or directory 之类的错误，请检查第三步是否做好。

imageio-2.2.0.tar.gz 解压后安装。./configure ，阅读readme。

Python 虚拟环境工具 virtualenv

sublime Text 解压后可使用，适用windows

Caffe源码，携带SSD，Examples，demo.py图像集测试,也可从Github源码下载，https://github.com/weiliu89/caffe/tree/ssd

IMageNet 图像集 标注工具 ，多边形标注，支持其他图像集标注。

咖啡2 Caffe2 Github20170803

咖啡二第三方依赖集合下载Caffe2_thirdparty

图像集 标注工具 LabelIMg ，sudo apt-get install pyqt4-dev-tools sudo pipinstall labelimg lableimg完成配置

再Caffe上使用 二进制文件 Mnist 图像 数据集

cs.toronto cifar10 图像集 境外下载慢 共4部分，最后一部分。

cifar10 图像集 第三部分

cs.toronto cifar10 图像集,共4部分

从www.cs.toronto.edu下载的文件，境外文件下载慢，提供给大家，放在Caffe根目录，data/cifar10直接使用

ubuntu 64 位

ubuntu 64位

bazel 64位 解压缩文件 双击既可安装

An-Overview-of-C++11-and-C++14

OpenCV的python版本。python3.8，支持所有x64位机器。使用匹配安装。windows推荐与python3.8配合使用。

解压到sln一级目录项目属性 C/C++ 附加包含目录 填写Libs/x86/opencv_v3.4.0/include路径 属性中链接器，所有选项附加库目录填写Libs/x86/opencv_v3.4.0/lib 附加依赖项： opencv_aruco340.lib;f.lib;opencv_bgsegm340.lib;opencv_bgsegm340d.lib;opencv_bioinspired340.lib;opencv_bioinspired340d.lib;opencv_calib3d340.lib;opencv_calib3d340d.lib;opencv_ccalib340.lib;opencv_ccalib340d.lib;opencv_core340.lib;opencv_core340d.lib;opencv_datasets340.lib;opencv_datasets340d.lib;opencv_dnn340.lib;opencv_dnn340d.lib;opencv_dpm340.lib;opencv_dpm340d.lib;opencv_face340.lib;opencv_face340d.lib;opencv_features2d340.lib;opencv_features2d340d.lib;opencv_flann340.lib;opencv_flann340d.lib;opencv_fuzzy340.lib;opencv_fuzzy340d.lib;opencv_highgui340.lib;opencv_highgui340d.lib;opencv_imgcodecs340.lib;opencv_imgcodecs340d.lib;opencv_imgproc340.lib;opencv_imgproc340d.lib;opencv_img_hash340.lib;opencv_img_hash340d.lib;opencv_line_descriptor340.lib;opencv_line_descriptor340d.lib;opencv_ml340.lib;opencv_ml340d.lib;opencv_objdetect340.lib;opencv_objdetect340d.lib;opencv_optflow340.lib;opencv_optflow340d.lib;opencv_phase_unwrapping340.lib;opencv_phase_unwrapping340d.lib;opencv_photo340.lib;opencv_photo340d.lib;opencv_plot340.lib;opencv_plot340d.lib;opencv_reg340.lib;opencv_reg340d.lib;opencv_rgbd340.lib;opencv_rgbd340d.lib;opencv_saliency340.lib;opencv_saliency340d.lib;opencv_shape340.lib;opencv_shape340d.lib;opencv_stereo340.lib;opencv_stereo340d.lib;opencv_stitching340.lib;opencv_stitching340d.lib;opencv_structured_light340.lib;opencv_structured_light340d.lib;opencv_superres340.lib;opencv_superres340d.lib;opencv_surface_matching340.lib;opencv_surface_matching340d.lib;opencv_text340.lib;opencv_text340d.lib;opencv_tracking340.lib;opencv_tracking340d.lib;opencv_video340.lib;opencv_video340d.lib;opencv_videoio340.lib;opencv_videoio340d.lib;opencv_videostab340.lib;opencv_videostab340d.lib;opencv_xfeatures2d340.lib;opencv_xfeatures2d340d.lib;opencv_ximgproc340.lib;opencv_ximgproc340d.lib;opencv_xobjdetect340.lib;opencv_xobjdetect340d.lib;opencv_xphoto340.lib;opencv_xphoto340d.lib;

tensorflow2.2.。使用pip安装。遇到无法下载的资源可以使用拷贝链接。再使用下载工具进行下载。本资源是官方正版，可到官方资源去下载。https://github.com/tensorflow/tensorflow/releases。

一个监测车辆跟踪的例子 import cv2 import numpy as np import math import matplotlib.pyplot as plt class Hog_descriptor(): def __init__(self, img, cell_size=16, bin_size=8): self.img = img self.img = np.sqrt(img / float(np.max(img))) self.img = self.img * 255 self.cell_size = cell_size self.bin_size = bin_size self.angle_unit = 360 / self.bin_size assert type(self.bin_size) == int, "bin_size should be integer," assert type(self.cell_size) == int, "cell_size should be integer," assert type(self.angle_unit) == int, "bin_size should be divisible by 360" def extract(self): height, width = self.img.shape gradient_magnitude, gradient_angle = self.global_gradient() gradient_magnitude = abs(gradient_magnitude) cell_gradient_vector = np.zeros((height / self.cell_size, width / self.cell_size, self.bin_size)) for i in range(cell_gradient_vector.shape[0]): for j in range(cell_gradient_vector.shape[1]): cell_magnitude = gradient_magnitude[i * self.cell_size:(i + 1) * self.cell_size, j * self.cell_size:(j + 1) * self.cell_size] cell_angle = gradient_angle[i * self.cell_size:(i + 1) * self.cell_size, j * self.cell_size:(j + 1) * self.cell_size] cell_gradient_vector[i][j] = self.cell_gradient(cell_magnitude, cell_angle) hog_image = self.render_gradient(np.zeros([height, width]), cell_gradient_vector) hog_vector = [] for i in range(cell_gradient_vector.shape[0] - 1): for j in range(cell_gradient_vector.shape[1] - 1): block_vector = [] block_vector.extend(cell_gradient_vector[i][j]) block_vector.extend(cell_gradient_vector[i][j + 1]) block_vector.extend(cell_gradient_vector[i + 1][j]) block_vector.extend(cell_gradient_vector[i + 1][j + 1]) mag = lambda vector: math.sqrt(sum(i ** 2 for i in vector)) magnitude = mag(block_vector) if magnitude != 0: normalize = lambda block_vector, magnitude: [element / magnitude for element in block_vector] block_vector = normalize(block_vector, magnitude) hog_vector.append(block_vector) return hog_vector, hog_image def global_gradient(self): gradient_values_x = cv2.Sobel(self.img, cv2.CV_64F, 1, 0, ksize=5) gradient_values_y = cv2.Sobel(self.img, cv2.CV_64F, 0, 1, ksize=5) gradient_magnitude = cv2.addWeighted(gradient_values_x, 0.5, gradient_values_y, 0.5, 0) gradient_angle = cv2.phase(gradient_values_x, gradient_values_y, angleInDegrees=True) return gradient_magnitude, gradient_angle def cell_gradient(self, cell_magnitude, cell_angle): orientation_centers = [0] * self.bin_size for i in range(cell_magnitude.shape[0]): for j in range(cell_magnitude.shape[1]): gradient_strength = cell_magnitude[i][j] gradient_angle = cell_angle[i][j] min_angle, max_angle, mod = self.get_closest_bins(gradient_angle) orientation_centers[min_angle] += (gradient_strength * (1 - (mod / self.angle_unit))) orientation_centers[max_angle] += (gradient_strength * (mod / self.angle_unit)) return orientation_centers def get_closest_bins(self, gradient_angle): idx = int(gradient_angle / self.angle_unit) mod = gradient_angle % self.angle_unit if idx == self.bin_size: return idx - 1, (idx) % self.bin_size, mod return idx, (idx + 1) % self.bin_size, mod def render_gradient(self, image, cell_gradient): cell_width = self.cell_size / 2 max_mag = np.array(cell_gradient).max() for x in range(cell_gradient.shape[0]): for y in range(cell_gradient.shape[1]): cell_grad = cell_gradient[x][y] cell_grad /= max_mag angle = 0 angle_gap = self.angle_unit for magnitude in cell_grad: angle_radian = math.radians(angle) x1 = int(x * self.cell_size + magnitude * cell_width * math.cos(angle_radian)) y1 = int(y * self.cell_size + magnitude * cell_width * math.sin(angle_radian)) x2 = int(x * self.cell_size - magnitude * cell_width * math.cos(angle_radian)) y2 = int(y * self.cell_size - magnitude * cell_width * math.sin(angle_radian)) cv2.line(image, (y1, x1), (y2, x2), int(255 * math.sqrt(magnitude))) angle += angle_gap return image img = cv2.imread('G://1.bmp', cv2.IMREAD_GRAYSCALE) hog = Hog_descriptor(img, cell_size=8, bin_size=8) vector, image = hog.extract() plt.imshow(image, cmap=plt.cm.gray) plt.show()

SVM的libsvm库。封装了libsvm，非常好的机器学习入门示例。可以用来做字符识别、是一个很好的模式识别例子。

字符模板，字符识别。字母A~Z，数字0~9的模板。OCR所需的字符模板。总共6393各字符。测试字符三百多个。

OPENCV4.2.0包括了non-free,可以做特征相关的算法。以下是程序。 #include "opencv2/opencv.hpp" void main() { cv::Mat lenaImg = cv::imread("G:\\1.bmp"); cv::Mat lenaImg2 = cv::imread("G:\\2.bmp"); auto akaze_detector = AKAZE::create(); vector<KeyPoint> kpts_01, kpts_02; Mat descriptors1, descriptors2; akaze_detector->detectAndCompute(lenaImg, Mat(), kpts_01, descriptors1); akaze_detector->detectAndCompute(lenaImg2, Mat(), kpts_02, descriptors2); // 定义描述子匹配 - 暴力匹配 Ptr<DescriptorMatcher> matcher = DescriptorMatcher::create(DescriptorMatcher::BRUTEFORCE); std::vector< DMatch > matches; matcher->match(descriptors1, descriptors2, matches); // 绘制匹配 Mat img_matches; drawMatches(lenaImg, kpts_01, lenaImg2, kpts_02, matches, img_matches); imshow("AKAZE-Matches", img_matches); imwrite("D://result.png", img_matches); waitKey(0); }

制作Mininst格式的数据，研究sklearn库等机器学习算法与库用得到。C++的VS2017测试通过。适用于车牌和数字识别。下面是Skearn测试程序。 #导入必备的包 import numpy as np import struct import matplotlib.pyplot as plt import os ##加载svm模型 from sklearn import svm ###用于做数据预处理 from sklearn import preprocessing import time #加载数据的路径 path='.' def load_mnist_train(path, kind='train'): labels_path = os.path.join(path,'%s-labels-idx1-ubyte'% kind) images_path = os.path.join(path,'%s-images-idx3-ubyte'% kind) with open(labels_path, 'rb') as lbpath: magic, n = struct.unpack('>II',lbpath.read(8)) labels = np.fromfile(lbpath,dtype=np.uint8) with open(images_path, 'rb') as imgpath: magic, num, rows, cols = struct.unpack('>IIII',imgpath.read(16)) images = np.fromfile(imgpath,dtype=np.uint8).reshape(len(labels), 784) return images, labels def load_mnist_test(path, kind='t10k'): labels_path = os.path.join(path,'%s-labels-idx1-ubyte'% kind) images_path = os.path.join(path,'%s-images-idx3-ubyte'% kind) with open(labels_path, 'rb') as lbpath: magic, n = struct.unpack('>II',lbpath.read(8)) labels = np.fromfile(lbpath,dtype=np.uint8) with open(images_path, 'rb') as imgpath: magic, num, rows, cols = struct.unpack('>IIII',imgpath.read(16)) images = np.fromfile(imgpath,dtype=np.uint8).reshape(len(labels), 784) return images, labels train_images,train_labels=load_mnist_train(path) test_images,test_labels=load_mnist_test(path) X=preprocessing.StandardScaler().fit_transform(train_images) X_train=X[0:60000] y_train=train_labels[0:60000] print(time.strftime('%Y-%m-%d %H:%M:%S')) model_svc = svm.LinearSVC() #model_svc = svm.SVC() model_svc.fit(X_train,y_train) print(time.strftime('%Y-%m-%d %H:%M:%S')) ##显示前30个样本的真实标签和预测值，用图显示 x=preprocessing.StandardScaler().fit_transform(test_images) x_test=x[0:10000] y_pred=test_labels[0:10000] print(model_svc.score(x_test,y_pred)) y=model_svc.predict(x) fig1=plt.figure(figsize=(8,8)) fig1.subplots_adjust(left=0,right=1,bottom=0,top=1,hspace=0.05,wspace=0.05) for i in range(100): ax=fig1.add_subplot(10,10,i+1,xticks=[],yticks=[]) ax.imshow(np.reshape(test_images[i], [28,28]),cmap=plt.cm.binary,interpolation='nearest') ax.text(0,2,"pred:"+str(y[i]),color='red') #ax.text(0,32,"real:"+str(test_labels[i]),color='blue') plt.show()

字符识别。安装tensorflow。 conda create -n cpu_avx2 python=3.7 The following packages will be downloaded: package | build ---------------------------|----------------- certifi-2020.4.5.1 | py37_0 156 KB openssl-1.1.1g | he774522_0 4.8 MB python-3.7.7 | h60c2a47_2 18.3 MB setuptools-46.1.3 | py37_0 528 KB sqlite-3.31.1 | h2a8f88b_1 1.3 MB zlib-1.2.11 | h62dcd97_4 132 KB ------------------------------------------------------------ Total: 25.2 MB The following NEW packages will be INSTALLED: ca-certificates pkgs/main/win-64::ca-certificates-2020.1.1-0 certifi pkgs/main/win-64::certifi-2020.4.5.1-py37_0 openssl pkgs/main/win-64::openssl-1.1.1g-he774522_0 pip pkgs/main/win-64::pip-20.0.2-py37_1 python pkgs/main/win-64::python-3.7.7-h60c2a47_2 setuptools pkgs/main/win-64::setuptools-46.1.3-py37_0 sqlite pkgs/main/win-64::sqlite-3.31.1-h2a8f88b_1 vc pkgs/main/win-64::vc-14.1-h0510ff6_4 vs2015_runtime pkgs/main/win-64::vs2015_runtime-14.16.27012-hf0eaf9b_1 wheel pkgs/main/win-64::wheel-0.34.2-py37_0 wincertstore pkgs/main/win-64::wincertstore-0.2-py37_0 zlib pkgs/main/win-64::zlib-1.2.11-h62dcd97_4 Proceed ([y]/n)? y Downloading and Extracting Packages setuptools-46.1.3 | 528 KB | ############################################################################ | 100% python-3.7.7 | 18.3 MB | ############################################################################ | 100% certifi-2020.4.5.1 | 156 KB | ############################################################################ | 100% zlib-1.2.11 | 132 KB | ###################

Minist原始数据，手写数字训练数据，含有60000个训练样本和10000测试数据。资源下载所需积分0 。或者去百度网盘免费下载。链接: https://pan.baidu.com/s/1YfjdwjqKromieGyZTprd7Q 提取码: q9ps

模仿mnist数据集制作自己的数据集代码。HOG+SVM，深度学习，识别中采用的数据集。可以适用多个平台，声明本代码在OPENCV3.3和VS2017测试通过。

$ gunzip < hdf5-X.Y.Z.tar.gz | tar xf - #解压缩 $ cd hdf5-X.Y.Z $ ./configure --prefix=/usr/local/hdf5 #安装路径 $ make $ make check # run test suite. $ make install $ make check-install # verify installation. 最后安装，支持h5编译。 sudo apt-get install libhdf5-serial-dev

VS,Linux Version,Extension:Clang,CMake,C/C++ IntelliSense.

三维重建。跟踪。ORB SLAM2 源码。

tensorflow 1.4版本 依赖库 python /home/suanfa/data/tf_depends/funcsigs-1.0.2.tar.gz /home/suanfa/data/tf_depends/funcsigs-1.0.2-py2.py3-none-any.whl /home/suanfa/data/tf_depends/mock-2.0.0.tar.gz /home/suanfa/data/tf_depends/mock-2.0.0-py2.py3-none-any.whl /home/suanfa/data/tf_depends/numpy-1.11.2.tar.gz /home/suanfa/data/tf_depends/pbr-3.1.1.tar.gz /home/suanfa/data/tf_depends/pbr-3.1.1-py2.py3-none-any.whl /home/suanfa/data/tf_depends/protobuf-3.1.0.post1.tar.gz /home/suanfa/data/tf_depends/protobuf-3.1.0.post1-py2.py3-none-any.whl /home/suanfa/data/tf_depends/setuptools-38.2.4.rar /home/suanfa/data/tf_depends/six-1.11.0.tar.gz /home/suanfa/data/tf_depends/six-1.11.0-py2.py3-none-any.whl /home/suanfa/data/tf_depends/tf_depends.rar

VOT2014竞赛测试集，Tracking测试集，共25个序列。 下载地址见附件。truck方法有很多，粒子滤波，频域方法。

VOC测试集和训练集，将下载的所有3个文件解压到一个文件夹。 下载地址见附件。VOC测试集和训练集，将下载的所有3个文件解压到一个文件夹。 学习跟踪可以从struck，频域方法研究。

caffe中提供了c++的接口，所以在c++矩阵对矩阵的处理是不可避免的，所以这里使用了eigen库来实现c++对矩阵、向量等的快速处理。 eigen是开源、并且不用编译的库，主要原因是它提供的实现都是模板，所以不能使用编译好的链接库。 下面介绍Ubuntu下的相关配置： 1、安装 该部分主要参照eigen3下载后的安装文档： 1）在INSTALL文件所在的文件路径新建一个文件夹如build_dir(m) 2）进入build_dir（cd build_dir） 3)cmake .. 4)make 5)sudo make install 安装后头文件在： usr/local/include/eigen3

进入下载文件夹，使用pip install pip-9.0.1-py2.py3-none-any.whl 安装

1）在安装OpenCV前需要安装的软件包有 GCC4.4.X or later，可通过命令sudo apt-get install build-essential安装 CMake2.6 or later SVN客户端 GTK+2.Xor higher, including headers(libgtk2.0-dev) pkgconfig Python2.6 or later and Numpy 1.5 or later with developerpackages(python-dev, python-numpy) ffmpegor libav development packages: libavcodec-dev, libavformat-dev,libswsacle-dev [可选]libdc13942.x [可选]libjpeg-dev,libpng-dev, libtiff-dev, libjasper-dev 所有的软件包都可在终端安装或者通过Synaptic软件管理器。 终端安装依赖项： sudo apt-get install build-essential libgtk2.0-dev libjpeg-dev libtiff4-dev libjasper-dev libopenexr-dev cmake python-dev python-numpy python-tk libtbb-dev libeigen2-dev yasm libfaac-dev libopencore-amrnb-dev libopencore-amrwb-dev libtheora-dev libvorbis-dev libxvidcore-dev libx264-dev libqt4-dev libqt4-opengl-dev sphinx-common texlive-latex-extra libv4l-dev libdc1394-22-dev libavcodec-dev libavformat-dev libswscale-dev 2）下载最新版OpenCV 打开网页：http://sourceforge.net/projects/opencvlibrary 下载安装包OpenCV-2.4.2.tar.bz2 sudo tar jxvf OpenCV-2.4.2.tar.bz2 -C /usr/local/ cd /usr/local/ sudo mv OpenCV-2.4.2 opencv cd opencv mkdir release cd release cmake -D WITH_TBB=ON -D BUILD_NEW_PYTHON_SUPPORT=ON -D WITH_V4L=ON -DINSTALL_C_EXAMPLES=ON -D INSTALL_PYTHON_EXAMPLES=ON -DBUILD_EXAMPLES=ON -D WITH_QT=ON -D WITH_OPENGL=ON .. make sudo make install 3）安装后的配置 添加库的路径 sudo gedit /etc/ld.so.conf.d/opencv.conf 添加内容 /usr/local/lib 在终端输入命令 sudo ldconfig 设置环境变量 sudo gedit /etc/bash.bashrc 在文件最后加入以下两行并保存： PKG_CONFIG_PATH=$PKG_CONFIG_PATH:/usr/local/lib/pkgconfig export PKG_CONFIG_PATH 此时重启Ubuntu或重新登录账户，使得OpenCV安装生效。 4）测试OpenCV自带例程 编译程序： cd /usr/local/opencv/samples/c chmod +x build_all.sh ./build_all.sh 运行程序： ./facedetect --cascade="/usr/local/share/OpenCV/haarcascades/haarcascade_frontalface_alt.xml" --scale=1.5 lena.jpg 运行结果可以看到脸部有圆框