由于該項目是針對中小學生競賽并且是第一次舉行，所以識別的目標交通標志僅僅只有直行、右轉、左轉和停車讓行。

數(shù)據(jù)集：https://pan.baidu.com/s/1sLl0CadEutv3PQXhmqpCXw 提取碼：mp2x

源代碼：https://github.com/ccxiao5/Traffic_sign_recognition

整體流程如下：

數(shù)據(jù)集收集（包括訓練集和測試集的分類）

圖像預處理

圖像標注

根據(jù)標注分割得到目標圖像

HOG特征提取

訓練得到模型

將模型帶入識別算法進行識別

我的數(shù)據(jù)目錄樹。其中test_images/train_images是收集得到原始數(shù)據(jù)集。realTest/realTrain是預處理后的圖像。dataTest/dataTrain是經(jīng)過分類處理得到的圖像，HogTest/HogTrain是通過XML標注后裁剪得到的圖像。HogTest_affine/HogTrain_affine是經(jīng)過仿射變換處理擴充的訓練集和測試集。imgTest_hog.txt/imgTrain_hog.txt是測試集和訓練集的Hog特征

一、圖像處理

由于得到的數(shù)據(jù)集圖像大小不一（如下），我們首先從中心區(qū)域裁剪并調整正方形圖像的大小，然后將處理后的圖像保存到realTrain和realTest里面。

圖片名稱對應關系如下：

img_label ={"000":"Speed_limit_5","001":"Speed_limit_15","002":"Speed_limit_30","003":"Speed_limit_40","004":"Speed_limit_50","005":"Speed_limit_60","006":"Speed_limit_70","007":"Speed_limit_80","008":"No straight or right turn","009":"No straight or left turn","010":"No straight","011":"No left turn","012":"Do not turn left and right","013":"No right turn","014":"No Overhead","015":"No U-turn","016":"No Motor vehicle","017":"No whistle","018":"Unrestricted speed_40","019":"Unrestricted speed_50","020":"Straight or turn right","021":"Straight","022":"Turn left","023":"Turn left or turn right","024":"Turn right","025":"Drive on the left side of the road","026":"Drive on the right side of the road","027":"Driving around the island","028":"Motor vehicle driving","029":"Whistle","030":"Non-motorized","031":"U-turn","032":"Left-right detour","033":"traffic light","034":"Drive cautiously","035":"Caution Pedestrians","036":"Attention non-motor vehicle","037":"Mind the children","038":"Sharp turn to the right","039":"Sharp turn to the left","040":"Downhill steep slope","041":"Uphill steep slope","042":"Go slow","044":"Right T-shaped cross","043":"Left T-shaped cross","045":"village","046":"Reverse detour","047":"Railway crossing-1","048":"construction","049":"Continuous detour","050":"Railway crossing-2","051":"Accident-prone road section","052":"stop","053":"No passing","054":"No Parking","055":"No entry","056":"Deceleration and concession","057":"Stop For Check"}

def center_crop(img_array, crop_size=-1, resize=-1, write_path=None):##從中心區(qū)域裁剪并調整正方形圖像的大小。

rows =img_array.shape[0]

cols= img_array.shape[1]if crop_size==-1 or crop_size>max(rows,cols):

crop_size=min(rows, cols)

row_s= max(int((rows-crop_size)/2), 0)

row_e= min(row_s+crop_size, rows)

col_s= max(int((cols-crop_size)/2), 0)

col_e= min(col_s+crop_size, cols)

img_crop=img_array[row_s:row_e,col_s:col_e,]if resize>0:

img_crop=cv2.resize(img_crop, (resize, resize))if write_path is notNone:

cv2.imwrite(write_path, img_crop)return img_crop

然后根據(jù)得到的realTrain和realTest自動生成帶有的xml文件

defwrite_img_to_xml(imgfile, xmlfile):

img=cv2.imread(imgfile)

img_folder, img_name=os.path.split(imgfile)

img_height, img_width, img_depth=img.shape

doc=Document()

annotation= doc.createElement("annotation")

doc.appendChild(annotation)

folder= doc.createElement("folder")

folder.appendChild(doc.createTextNode(img_folder))

annotation.appendChild(folder)

filename= doc.createElement("filename")

filename.appendChild(doc.createTextNode(img_name))

annotation.appendChild(filename)

size= doc.createElement("size")

annotation.appendChild(size)

width= doc.createElement("width")

width.appendChild(doc.createTextNode(str(img_width)))

size.appendChild(width)

height= doc.createElement("height")

height.appendChild(doc.createTextNode(str(img_height)))

size.appendChild(height)

depth= doc.createElement("depth")

depth.appendChild(doc.createTextNode(str(img_depth)))

size.appendChild(depth)

with open(xmlfile,"w") as f:

doc.writexml(f, indent="\t", addindent="\t", newl="\n", encoding="utf-8")

/home/xiao5/Desktop/Test2/data/realTest/PNGImages

000_1_0001_1_j.png

640

640

3

然后對realTrain和realTest的圖片進行標注，向默認XML添加新的信息（矩形信息）。

PNGImages

021_1_0001_1_j.png

C:\Users\xiao5\Desktop\realTest\PNGImages\021_1_0001_1_j.png

Unknown

640

640

3

0

Straight

Unspecified

0

0

13

22

573

580

處理完后利用我們添加的矩形將圖片裁剪下來并且重命名進行分類。主要思路是：解析XML文檔，根據(jù)標簽進行分類，如果是直行、右轉、左轉、停止，那么就把它從原圖中裁剪下來并重命名,如果沒有那么就認為是負樣本，其中在處理負樣本的時候，我進行了顏色識別，把一張負樣本圖片根據(jù)顏色（紅色、藍色）裁剪成幾張負樣本，這樣做的好處是：我們在進行交通標志的識別時，也是使用的顏色識別來選取到交通標志，我們從負樣本中分割出來的相近顏色樣本有利于負樣本的訓練，提高模型精度。

def produce_proposals(xml_dir, write_dir, square=False, min_size=30):##返回proposal_num對象

proposal_num ={}for cls_name inclasses_name:

proposal_num[cls_name]=0

index =0for xml_file inos.listdir(xml_dir):

img_path, labels=parse_xml(os.path.join(xml_dir,xml_file))

img=cv2.imread(img_path)##如果圖片中沒有出現(xiàn)定義的那幾種交通標志就把它當成負樣本

if len(labels) ==0:

neg_proposal_num= produce_neg_proposals(img_path, write_dir, min_size, square, proposal_num["background"])

proposal_num["background"] =neg_proposal_numelse:

proposal_num= produce_pos_proposals(img_path, write_dir, labels, min_size, square=True, proposal_num=proposal_num)if index%100 ==0:print ("total xml file number =", len(os.listdir(xml_dir)), "current xml file number =", index)print ("proposal num =", proposal_num)

index+= 1

return proposal_num

為了提高模型的精確度，還對目標圖片（四類圖片）進行仿射變換來擴充訓練集。

defaffine(img, delta_pix):

rows, cols, _=img.shape

pts1=np.float32([[0,0], [rows,0], [0, cols]])

pts2= pts1 +delta_pix

M=cv2.getAffineTransform(pts1, pts2)

res=cv2.warpAffine(img, M, (rows, cols))returnresdefaffine_dir(img_dir, write_dir, max_delta_pix):

img_names=os.listdir(img_dir)

img_names= [img_name for img_name in img_names if img_name.split(".")[-1]=="png"]for index, img_name inenumerate(img_names):

img=cv2.imread(os.path.join(img_dir,img_name))

save_name= os.path.join(write_dir, img_name.split(".")[0]+"f.png")

delta_pix= np.float32(np.random.randint(-max_delta_pix,max_delta_pix+1,[3,2]))

img_a=affine(img, delta_pix)

cv2.imwrite(save_name, img_a)

二、HOG特征提取

處理好圖片后分別對訓練集和測試集進行特征提取得到imgTest_HOG.txt和imgTrain_HOG.txt

def hog_feature(img_array, resize=(64,64)):##提取HOG特征

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

img=cv2.resize(img, resize)

bins= 9cell_size= (8, 8)

cpb= (2, 2)

norm= "L2"features= ft.hog(img, orientations=bins, pixels_per_cell=cell_size,

cells_per_block=cpb, block_norm=norm, transform_sqrt=True)returnfeaturesdef extra_hog_features_dir(img_dir, write_txt, resize=(64,64)):##提取目錄中所有圖像HOG特征

img_names=os.listdir(img_dir)

img_names= [os.path.join(img_dir, img_name) for img_name inimg_names]ifos.path.exists(write_txt):

os.remove(write_txt)

with open(write_txt,"a") as f:

index=0for img_name inimg_names:

img_array=cv2.imread(img_name)

features=hog_feature(img_array, resize)

label_name= img_name.split("/")[-1].split("_")[0]

label_num=img_label[label_name]

row_data= img_name + "\t" + str(label_num) + "\t"

for element infeatures:

row_data= row_data + str(round(element,3)) + " "row_data= row_data + "\n"f.write(row_data)if index%100 ==0:print ("total image number =", len(img_names), "current image number =", index)

index+= 1

三、模型訓練

利用得到的HOG特征進行訓練模型得到svm_model.pkl

defload_hog_data(hog_txt):

img_names=[]

labels=[]

hog_features=[]

with open(hog_txt,"r") as f:

data=f.readlines()for row_data indata:

row_data=row_data.rstrip()

img_path, label, hog_str= row_data.split("\t")

img_name= img_path.split("/")[-1]

hog_feature= hog_str.split(" ")

hog_feature= [float(hog) for hog inhog_feature]#print "hog feature length = ", len(hog_feature)

img_names.append(img_name)

labels.append(label)

hog_features.append(hog_feature)returnimg_names, np.array(labels), np.array(hog_features)def svm_train(hog_features, labels, save_path="./svm_model.pkl"):

clf= SVC(C=10, tol=1e-3, probability =True)

clf.fit(hog_features, labels)

joblib.dump(clf, save_path)print ("finished.")

四、交通標志識別及實驗測試

交通標志識別的流程：顏色識別得到閾值范圍內的二值圖、然后進行輪廓識別、剔除多余矩陣。

defpreprocess_img(imgBGR):##將圖像由RGB模型轉化成HSV模型

imgHSV =cv2.cvtColor(imgBGR, cv2.COLOR_BGR2HSV)

Bmin= np.array([110, 43, 46])

Bmax= np.array([124, 255, 255])##使用inrange(HSV,lower,upper)設置閾值去除背景顏色

img_Bbin =cv2.inRange(imgHSV,Bmin, Bmax)

Rmin2= np.array([165, 43, 46])

Rmax2= np.array([180, 255, 255])

img_Rbin=cv2.inRange(imgHSV,Rmin2, Rmax2)

img_bin=np.maximum(img_Bbin, img_Rbin)returnimg_bin'''提取輪廓,返回輪廓矩形框'''

def contour_detect(img_bin, min_area=0, max_area=-1, wh_ratio=2.0):

rects=[]##檢測輪廓，其中cv2.RETR_EXTERNAL只檢測外輪廓，cv2.CHAIN_APPROX_NONE 存儲所有的邊界點

##findContours返回三個值:第一個值返回img，第二個值返回輪廓信息，第三個返回相應輪廓的關系

contours, hierarchy=cv2.findContours(img_bin.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)if len(contours) ==0:returnrects

max_area= img_bin.shape[0]*img_bin.shape[1] if max_area<0 elsemax_areafor contour incontours:

area=cv2.contourArea(contour)if area >= min_area and area <=max_area:

x, y, w, h=cv2.boundingRect(contour)if 1.0*w/h < wh_ratio and 1.0*h/w

rects.append([x,y,w,h])return rects

然后加載模型進行測驗

if __name__ == "__main__":

cap=cv2.VideoCapture(0)

cv2.namedWindow('camera')

cv2.resizeWindow("camera",640,480)

cols=int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))

rows=int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))

clf= joblib.load("/home/xiao5/Desktop/Test2/svm_model.pkl")

i=0while (1):

i+=1ret, img=cap.read()

img_bin=preprocess_img(img)

min_area= img_bin.shape[0]*img.shape[1]/(25*25)

rects= contour_detect(img_bin, min_area=min_area)ifrects:

Max_X=0

Max_Y=0

Max_W=0

Max_H=0for r inrects:if r[2]*r[3]>=Max_W*Max_H:

Max_X,Max_Y,Max_W,Max_H=r

proposal= img[Max_Y:(Max_Y+Max_H),Max_X:(Max_X+Max_W)]##用Numpy數(shù)組對圖像像素進行訪問時，應該先寫圖像高度所對應的坐標(y,row)，再寫圖像寬度對應的坐標(x,col)。

cv2.rectangle(img,(Max_X,Max_Y), (Max_X+Max_W,Max_Y+Max_H), (0,255,0), 2)

cv2.imshow("proposal", proposal)

cls_prop=hog_extra_and_svm_class(proposal, clf)

cls_prop= np.round(cls_prop, 2)

cls_num= np.argmax(cls_prop)##找到最大相似度的索引

if cls_names[cls_num] is not "background":print(cls_names[cls_num])else:print("N/A")

cv2.imshow('camera',img)

cv2.waitKey(40)

cv2.destroyAllWindows()

cap.release()

總結

以上是生活随笔為你收集整理的python识别简单训练模型_Python3+OpenCV实现简单交通标志识别的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內容還不錯，歡迎將生活随笔推薦給好友。