svm对mnist分类matlab,计算机视觉---手写体识别，SVM分类

这周三robomasters的大佬要纳新了，还说让带着博客去，赶紧来更新一期(为了大佬看着方便，我给每一行都加上了注释，大佬且慢端详)，考虑了上次的KNN属于慵懒机器学习算法且准确率还那么低，这次换成SVM试一试。

我没有选用sk-learn的手写体数据集的原因是他的数据集每张图片是8×8的，而我下载的mnist的digits是一张包含5000张手写体图片的1000×2000的图片，也就是每张手写体图片是20×20的，心想也许这样的准确率会比较好。

digits.png

这是这次的效果图，准确率也不是很好。

效果图

这是该程序的准确率和初始化时间。

运行截图

用MNIST自己的数据测试的准确率很高，可是测自己的手写体准确率很低，好多人都有这个问题，可能是写字的笔不一样吧。。。

看程序。

S.1 划分数据集

def initSvm():

img = cv2.imread('digits.png') #读取数据集

gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) #转灰度图

cells = [np.hsplit(row,100) for row in np.vsplit(gray,50)] #划分数据集

train = np.array(cells).reshape(-1,400).astype(np.float32) #将图片转为行向量

trainLabel = np.repeat(np.arange(10),500) #建立索引

return train, trainLabel

S.2 寻找数字位置，并预测数值

def findRoi(frame, thresValue):

rois = []

gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

gray2 = cv2.dilate(gray,None,iterations=2) #两次膨胀

gray2 = cv2.erode(gray2,None,iterations=2) #两次腐蚀

edges = cv2.absdiff(gray,gray2) #做差，建立sobel算子进行边缘检测

x = cv2.Sobel(edges,cv2.CV_16S,1,0)

y = cv2.Sobel(edges,cv2.CV_16S,0,1)

absX = cv2.convertScaleAbs(x)

absY = cv2.convertScaleAbs(y)

dst = cv2.addWeighted(absX,0.5,absY,0.5,0)

ret, ddst = cv2.threshold(dst,thresValue,255,cv2.THRESH_BINARY) #转为二值图

im, contours, hierarchy = cv2.findContours(ddst,cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE) #寻找边界

for c in contours:

x, y, w, h = cv2.boundingRect(c)

if w > 10 and h > 20:

rois.append((x,y,w,h))

digits = []

for r in rois:

x, y, w, h = r

digit= findDigit(edges[y:y+h,x:x+w], 50) #预测数值

#digits.append(cv2.resize(th,(20,20)))

cv2.rectangle(frame, (x,y), (x+w,y+h), (153,153,0), 2) #绘制矩形框

cv2.putText(frame, str(digit), (x,y), cv2.FONT_HERSHEY_SIMPLEX, 1, (127,0,255), 2) #绘制预测数字

return edges

def findDigit(roi, thresValue):

ret, th = cv2.threshold(roi, thresValue, 255, cv2.THRESH_BINARY)

th = cv2.resize(th,(20,20)) #图片转为20×20

out = th.reshape(-1,400).astype(np.float32) #图片转为行向量

result = lsvc.predict(out) #预测结果

return result

S.3 训练模型，输出初始化时间及准确率

X, Y = initSvm()

X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size = 0.25, random_state = 0)

from sklearn.preprocessing import StandardScaler

from sklearn.svm import LinearSVC

ss = StandardScaler() #数据标准化

X_train = ss.fit_transform(X_train)

X_test = ss.transform(X_test)

lsvc = LinearSVC()

lsvc.fit(X_train, y_train) #训练模型

y_predict = lsvc.predict(X_test) #预测测试数据集

print 'The Accuracy of Linear SVC is: ', lsvc.score(X_test, y_test) #获得评分

endtime = datetime.datetime.now()

print 'The time of SVM init is: ', (endtime - starttime).seconds, 's' #计算初始化时间

S.4 打开摄像头，开始吧！

cap = cv2.VideoCapture(0)

width = 426*2

height = 480

videoFrame = cv2.VideoWriter('frame.avi',cv2.VideoWriter_fourcc('M','J','P','G'),25,(int(width),int(height)),True)

while True:

ret, frame = cap.read()

frame = frame[:,:426]

edges = findRoi(frame, 50)

newEdges = cv2.cvtColor(edges, cv2.COLOR_GRAY2BGR)

newFrame = np.hstack((frame,newEdges))

cv2.imshow('frame', newFrame)

videoFrame.write(newFrame) #保存视频

key = cv2.waitKey(1) & 0xff

if key == ord('q'):

break

完整代码如下

#!/usr/bin/python3

# -*- coding: UTF-8 -*-

import cv2

import numpy as np

from sklearn.cross_validation import train_test_split

import datetime

starttime = datetime.datetime.now()

def initSvm():

img = cv2.imread('digits.png')

gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

cells = [np.hsplit(row,100) for row in np.vsplit(gray,50)]

train = np.array(cells).reshape(-1,400).astype(np.float32)

trainLabel = np.repeat(np.arange(10),500)

return train, trainLabel

def findRoi(frame, thresValue):

rois = []

gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

gray2 = cv2.dilate(gray,None,iterations=2)

gray2 = cv2.erode(gray2,None,iterations=2)

edges = cv2.absdiff(gray,gray2)

x = cv2.Sobel(edges,cv2.CV_16S,1,0)

y = cv2.Sobel(edges,cv2.CV_16S,0,1)

absX = cv2.convertScaleAbs(x)

absY = cv2.convertScaleAbs(y)

dst = cv2.addWeighted(absX,0.5,absY,0.5,0)

ret, ddst = cv2.threshold(dst,thresValue,255,cv2.THRESH_BINARY)

im, contours, hierarchy = cv2.findContours(ddst,cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

for c in contours:

x, y, w, h = cv2.boundingRect(c)

if w > 10 and h > 20:

rois.append((x,y,w,h))

digits = []

for r in rois:

x, y, w, h = r

digit= findDigit(edges[y:y+h,x:x+w], 50)

#digits.append(cv2.resize(th,(20,20)))

cv2.rectangle(frame, (x,y), (x+w,y+h), (153,153,0), 2)

cv2.putText(frame, str(digit), (x,y), cv2.FONT_HERSHEY_SIMPLEX, 1, (127,0,255), 2)

return edges

def findDigit(roi, thresValue):

ret, th = cv2.threshold(roi, thresValue, 255, cv2.THRESH_BINARY)

th = cv2.resize(th,(20,20))

out = th.reshape(-1,400).astype(np.float32)

result = lsvc.predict(out)

return result

X, Y = initSvm()

X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size = 0.25, random_state = 0)

from sklearn.preprocessing import StandardScaler

from sklearn.svm import LinearSVC

ss = StandardScaler()

X_train = ss.fit_transform(X_train)

X_test = ss.transform(X_test)

lsvc = LinearSVC()

lsvc.fit(X_train, y_train)

y_predict = lsvc.predict(X_test)

print 'The Accuracy of Linear SVC is: ', lsvc.score(X_test, y_test)

endtime = datetime.datetime.now()

print 'The time of SVM init is: ', (endtime - starttime).seconds, 's'

cap = cv2.VideoCapture(0)

width = 426*2

height = 480

videoFrame = cv2.VideoWriter('frame.avi',cv2.VideoWriter_fourcc('M','J','P','G'),25,(int(width),int(height)),True)

while True:

ret, frame = cap.read()

frame = frame[:,:426]

edges = findRoi(frame, 50)

newEdges = cv2.cvtColor(edges, cv2.COLOR_GRAY2BGR)

newFrame = np.hstack((frame,newEdges))

cv2.imshow('frame', newFrame)

videoFrame.write(newFrame)

key = cv2.waitKey(1) & 0xff

if key == ord('q'):

break