(1)数据准备
数据集介绍:
数据集中存放的是1223幅图像,其中756个负样本(图像名称为0.1~0.756),458个正样本(图像名称为1.1~1.458),其中:"."前的标号为样本标签,"."后的标号为样本序号
(2)利用python读取文件夹中所有图像
'''
Load the image files form the folder
input:
imgDir: the direction of the folder
imgName:the name of the folder
output:
data:the data of the dataset
label:the label of the datset
'''
def load_Img(imgDir,imgFoldName):
imgs = os.listdir(imgDir+imgFoldName)
imgNum = len(imgs)
data = np.empty((imgNum,1,12,12),dtype="float32")
label = np.empty((imgNum,),dtype="uint8")
for i in range (imgNum):
img = Image.open(imgDir+imgFoldName+"/"+imgs[i])
arr = np.asarray(img,dtype="float32")
data[i,:,:,:] = arr
label[i] = int(imgs[i].split('.')[0])
return data,label
这里得到的data和label都是ndarray数据
data: (1223,1,12,12)
label:(1223,)
注:nddary数据类型是numpy提供的一个数据类型,即N-dimensional array,它弥补了python中array不支持多维的缺陷
(3)调用方式
craterDir = "./data/CraterImg/Adjust/"
foldName = "East_CraterAdjust12"
data, label = load_Img(craterDir,foldName)