检测与实例分割学习笔记

版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/yunxinan/article/details/83214177

#第一部分
import numpy as np
def rad(x):
    return x * np.pi / 180

img = cv2.imread("C:/Users/Administrator/Desktop/1010test/21.jpg")
#cv2.imshow("original", img)

img = cv2.copyMakeBorder(img, 200, 200, 200, 200, cv2.BORDER_CONSTANT, 0)
w, h = img.shape[0:2]

anglex =-10
angley = 0
anglez = 0
fov = 42
while 1:
    # 镜头与图像间的距离，21为半可视角，算z的距离是为了保证在此可视角度下恰好显示整幅图像
    z = np.sqrt(w ** 2 + h ** 2) / 2 / np.tan(rad(fov / 2))
    # 齐次变换矩阵
    rx = np.array([[1, 0, 0, 0],
                   [0, np.cos(rad(anglex)), -np.sin(rad(anglex)), 0],
                   [0, -np.sin(rad(anglex)), np.cos(rad(anglex)), 0, ],
                   [0, 0, 0, 1]], np.float32)

    ry = np.array([[np.cos(rad(angley)), 0, np.sin(rad(angley)), 0],
                   [0, 1, 0, 0],
                   [-np.sin(rad(angley)), 0, np.cos(rad(angley)), 0, ],
                   [0, 0, 0, 1]], np.float32)

    rz = np.array([[np.cos(rad(anglez)), np.sin(rad(anglez)), 0, 0],
                   [-np.sin(rad(anglez)), np.cos(rad(anglez)), 0, 0],
                   [0, 0, 1, 0],
                   [0, 0, 0, 1]], np.float32)

    r = rx.dot(ry).dot(rz)

    # 四对点的生成
    pcenter = np.array([h / 2, w / 2, 0, 0], np.float32)

    p1 = np.array([0, 0, 0, 0], np.float32) - pcenter
    p2 = np.array([w, 0, 0, 0], np.float32) - pcenter
    p3 = np.array([0, h, 0, 0], np.float32) - pcenter
    p4 = np.array([w, h, 0, 0], np.float32) - pcenter

    dst1 = r.dot(p1)
    dst2 = r.dot(p2)
    dst3 = r.dot(p3)
    dst4 = r.dot(p4)

    list_dst = [dst1, dst2, dst3, dst4]

    org = np.array([[0, 0],
                    [w, 0],
                    [0, h],
                    [w, h]], np.float32)

    dst = np.zeros((4, 2), np.float32)

    # 投影至成像平面
    for i in range(4):
        dst[i, 0] = list_dst[i][0] * z / (z - list_dst[i][2]) + pcenter[0]
        dst[i, 1] = list_dst[i][1] * z / (z - list_dst[i][2]) + pcenter[1]

    warpR = cv2.getPerspectiveTransform(org, dst)

    result = cv2.warpPerspective(img, warpR, (h, w))
    cv2.imshow("result", result)
    c = cv2.waitKey(0)

    # anglex += 3            #auto rotate
    # anglez += 1             #auto rotate
    # angley += 2            #auto rotate

    # 键盘控制
    if 27 == c:  # Esc quit
        break;
    if c == ord('w'):
        anglex += 1
    if c == ord('s'):
        anglex -= 1
    if c == ord('a'):
        angley += 1
        # dx=0
    if c == ord('d'):
        angley -= 1
    if c == ord('u'):
        anglez += 1
    if c == ord('p'):
        anglez -= 1
    if c == ord('t'):
        fov += 1
    if c == ord('r'):
        fov -= 1
    if c == ord(' '):
        anglex = angley = anglez = 0
    if c == ord('q'):
        print("==============")
        print('旋转矩阵：\n', r)
        print("angle alpha: ", anglex, 'angle beta: ', angley, "dz: ", anglez, ": ", z)

cv2.waitKey(0)
cv2.destroyAllWindows()
#第二部分
import cv2
import numpy as np


def get_image(path):
    #获取图片
    img=cv2.imread(path)
    gray=cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)

    return img, gray

def Gaussian_Blur(gray):
    # 高斯去噪
    blurred = cv2.GaussianBlur(gray, (9, 9),0)

    return blurred

def Sobel_gradient(blurred):
    # 索比尔算子来计算x、y方向梯度
    gradX = cv2.Sobel(blurred, ddepth=cv2.CV_32F, dx=1, dy=0)
    gradY = cv2.Sobel(blurred, ddepth=cv2.CV_32F, dx=0, dy=1)

    gradient = cv2.subtract(gradX, gradY)
    gradient = cv2.convertScaleAbs(gradient)

    return gradX, gradY, gradient

def Thresh_and_blur(gradient):

    blurred = cv2.GaussianBlur(gradient, (9, 9),0)
    (_, thresh) = cv2.threshold(blurred, 90, 255, cv2.THRESH_BINARY)

    return thresh

def image_morphology(thresh):
    # 建立一个椭圆核函数
    kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (25, 25))
    # 执行图像形态学, 细节直接查文档，很简单
    closed = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)
    closed = cv2.erode(closed, None, iterations=4)
    closed = cv2.dilate(closed, None, iterations=4)

    return closed

def findcnts_and_box_point(closed):
    # 这里opencv3返回的是三个参数
    (_, cnts, _) = cv2.findContours(closed.copy(),
        cv2.RETR_LIST,
        cv2.CHAIN_APPROX_SIMPLE)
    c = sorted(cnts, key=cv2.contourArea, reverse=True)[0]
    # compute the rotated bounding box of the largest contour
    rect = cv2.minAreaRect(c)
    box = np.int0(cv2.boxPoints(rect))

    return box

def drawcnts_and_cut(original_img, box):
    # 因为这个函数有极强的破坏性，所有需要在img.copy()上画
    # draw a bounding box arounded the detected barcode and display the image
    draw_img = cv2.drawContours(original_img.copy(), [box], -1, (0, 0, 255), 3)

    Xs = [i[0] for i in box]
    Ys = [i[1] for i in box]
    x1 = min(Xs)
    x2 = max(Xs)
    y1 = min(Ys)
    y2 = max(Ys)
    hight = y2 - y1
    width = x2 - x1
    crop_img = original_img[y1:y1+hight, x1:x1+width]

    return draw_img, crop_img

def walk():

    img_path = r'C:/Users/Administrator/Desktop/1010test/3.png'
    save_path = r'C:/Users/Administrator/Desktop/1010test/3_save.png'
    original_img, gray = get_image(img_path)
    blurred = Gaussian_Blur(gray)
    gradX, gradY, gradient = Sobel_gradient(blurred)
    thresh = Thresh_and_blur(gradient)
    closed = image_morphology(thresh)
    box = findcnts_and_box_point(closed)
    draw_img, crop_img = drawcnts_and_cut(original_img,box)

    # 暴力一点，把它们都显示出来看看

    cv2.imshow('original_img', original_img)
    cv2.imshow('blurred', blurred)
    cv2.imshow('gradX', gradX)
    cv2.imshow('gradY', gradY)
    cv2.imshow('final', gradient)
    cv2.imshow('thresh', thresh)
    cv2.imshow('closed', closed)
    cv2.imshow('draw_img', draw_img)
    cv2.imshow('crop_img', crop_img)
    cv2.waitKey(20171219)
    cv2.imwrite(save_path, crop_img)

walk()
#第三部分
from PIL import Image
def ResizeImage(filein,fileout,width,height,type):
    img = Image.open(filein)
    out = img.resize((width,height),Image.ANTIALIAS)
    out.save(fileout,type)
if __name__ =="__main__":
    filein = r'C:/Users/Administrator/Desktop/1010test/15_dst.jpg'
    fileout = r'C:/Users/Administrator/Desktop/1010test/15_new.jpg'
    width = 440
    height = 300
    type = 'png'
    ResizeImage(filein, fileout, width, height, type)
#第四部分
from PIL import Image
# 1
im = Image.open('D:/STRPIC/6.jpg')
img_size = im.size
print("证号{}".format(img_size))
left = 120
upper = 16
right = 520
lower = 57
region = im.crop((left,upper,right,lower))
region.save("C:/Users/Administrator/Desktop/test/Crop_t1.jpg")
# 2
img_size = im.size
print("姓名{}".format(img_size))
left = 180
upper = 75
right = 485
lower = 115
region = im.crop((left,upper,right,lower))
region.save("C:/Users/Administrator/Desktop/test/Crop_t2.jpg")
# 3
img_size = im.size
print("性别{}".format(img_size))
left = 30
upper = 114
right = 600
lower = 150
region = im.crop((left,upper,right,lower))
region.save("C:/Users/Administrator/Desktop/test/Crop_t3.jpg")
# 4
img_size = im.size
print("住址{}".format(img_size))
left = 30
upper = 165
right = 600
lower = 200
region = im.crop((left,upper,right,lower))
region.save("C:/Users/Administrator/Desktop/test/Crop_t4.jpg")
# 5
img_size = im.size
print("出生日期{}".format(img_size))
left = 162
upper = 245
right = 450
lower = 290
region = im.crop((left,upper,right,lower))
region.save("C:/Users/Administrator/Desktop/test/Crop_t5.jpg")
# 6
img_size = im.size
print("初次领证日期{}".format(img_size))
left = 162
upper = 295
right = 450
lower = 340
region = im.crop((left,upper,right,lower))
region.save("C:/Users/Administrator/Desktop/test/Crop_t6.jpg")
# 7
img_size = im.size
print("准驾车型{}".format(img_size))
left = 162
upper = 345
right = 450
lower = 400
region = im.crop((left,upper,right,lower))
region.save("C:/Users/Administrator/Desktop/test/Crop_t7.jpg")
# 8
img_size = im.size
print("有效日期{}".format(img_size))
left = 30
upper = 390
right = 450
lower = 435
region = im.crop((left,upper,right,lower))
region.save("C:/Users/Administrator/Desktop/test/Crop_t8.jpg")

master R-cnn
http://www.sohu.com/a/228409487_129720
http://www.ijiandao.com/2b/baijia/95468.html
https://blog.csdn.net/u011974639/article/details/79806893

https://blog.csdn.net/weixin_42615068/article/details/82466454
https://blog.csdn.net/amusi1994/article/details/82356417
https://blog.csdn.net/hw5226349/article/details/81906882
https://blog.csdn.net/u011974639/article/details/79595179
http://www.sohu.com/a/245563314_651893
非常重要
https://blog.csdn.net/xiao__run/article/details/80393016?utm_source=blogxgwz3
https://github.com/sfzhang15/RefineDet
https://github.com/msracver/Relation-Networks-for-Object-Detection
https://github.com/msracver/Relation-Networks-for-Object-Detection