def image_corrosion(img_src, img_dst):
src = read_img(img_src)
## b.设置卷积核5*5
kernel = np.ones((5, 5), np.uint8)
## c.图像的腐蚀，默认迭代次数
erosion = cv2.erode(src, kernel)
#腐蚀后，图像明显变黑很多
cv2.imwrite(img_dst, erosion)
def image_sharpen(img_src, img_dst):
"""
:param img_src:传入图片的路径
:param img_dst:模糊化后图片的路径
:return:
"""
image = read_img(img_src)
kernel = np.array([[0, -1, 0], [-1, 5, -1], [0, -1, 0]], np.float32) # 锐化
dst = cv2.filter2D(image, -1, kernel=kernel) #滤波器
cv2.imwrite(img_dst, dst)
def gray_scale(img_src_path, img_dst_path):
"""
:param img_src_path:图片的源路径
:param img_dst_path:灰度化的图片路径
:return:img_dst_path
"""
img = read_img(img_src_path)
gray = cv2.cvtColor(
img, cv2.COLOR_BGR2GRAY) #RGB颜色空间以R(Red:红)、G(Green:绿)、B(Blue:蓝)
cv2.imwrite(img_dst_path, gray)
return img_dst_path
def image_dilate(img_src, img_dst):
#膨胀类似与 '领域扩张' ,将图像的高亮区域或白色部分进行扩张,其运行结果图比原图的高亮区域更大
src = read_img(img_src)
## 设置卷积核5*5
kernel = np.ones((5, 5), np.uint8)
## 图像的腐蚀，默认迭代次数
erosion = cv2.erode(src, kernel)
## 图像的膨胀
dst = cv2.dilate(erosion,
kernel) #在图像上进行从左到右，从上到下的平移，如果方框中存在白色，那么这个方框内所有的颜色都是白色
cv2.imwrite(img_dst, dst)
def image_blur(img_src, img_dst):
"""
:param img_src:传入图片的路径
:param img_dst:模糊化后图片的路径
:return:
"""
src = read_img(img_src)
dst = cv2.blur(src, (15, 1)) #均值滤波
# cv2.imshow("blur_demo", dst)
cv2.imwrite(img_dst, dst)
def binary_local_threshold(img_src,img_dst):
"""
:param img_src:传入图片的路径
:param img_dst:模糊化后图片的路径
:return:
"""
src = read_img(img_src)
gray = cv2.cvtColor(src, cv2.COLOR_RGB2GRAY) #把输入图像灰度化
#自适应阈值化能够根据图像不同区域亮度分布，改变阈值
binary = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C,cv2.THRESH_BINARY, 25, 10)
# cv2.namedWindow("binary1", cv2.WINDOW_NORMAL)
# cv2.imshow("binary1", binary)
cv2.imwrite(img_dst, binary)
output)
return output
def gasuss_noise(image, mean=0, var=0.001):
'''
添加高斯噪声
mean : 均值
var : 方差
'''
image = np.array(image / 255, dtype=float)
noise = np.random.normal(mean, var**0.5, image.shape)
out = image + noise
if out.min() < 0:
low_clip = -1.
else:
low_clip = 0.
out = np.clip(out, low_clip, 1.0)
out = np.uint8(out * 255)
cv2.imwrite(
"D:/pythonProject/ImageRecognitionAndClassification/UnitTest/ImgReco/small_tool_6.png",
out)
return out
img_path = "D:/pythonProject/ImageRecognitionAndClassification/UnitTest/ImgReco/small_tool_2.png"
#gasuss_noise(image=read_img(img_path))
sp_noise(image=read_img(img_path), prob=0.1)