1. 影像梯度

1.1 Soble算子

• 求x,y方向影像梯度

import cv2 as cv

def sobel_demo(image):
    grad_x = cv.Sobel(image, cv.CV_32F, 1, 0)	# 求x方向的梯度，對x求一階導數
    grad_y = cv.Sobel(image, cv.CV_32F, 0, 1)	# 求x方向的梯度，對y求一階導數
    gradx = cv.convertScaleAbs(grad_x)
    grady = cv.convertScaleAbs(grad_y)
    cv.imshow("grandient-x", gradx)
    cv.imshow("grandient-y", grady)

image = cv.resize(cv.imread("01.jpg"), (300,400))
cv.imshow("image", image)
sobel_demo(image)
cv.waitKey(0)
cv.destroyAllWindows()

• 求整體梯度

import cv2 as cv

def sobel_demo(image):
    grad_x = cv.Sobel(image, cv.CV_32F, 1, 0)
    grad_y = cv.Sobel(image, cv.CV_32F, 0, 1)
    gradx = cv.convertScaleAbs(grad_x)
    grady = cv.convertScaleAbs(grad_y)
    gradxy = cv.addWeighted(gradx, 0.5, grady, 0.5, 0)
    cv.imshow("gradxy",gradxy)

image = cv.resize(cv.imread("01.jpg"), (300,400))
cv.imshow("image", image)
sobel_demo(image)
cv.waitKey(0)
cv.destroyAllWindows()

1.2 拉普拉斯算子

• 拉普拉斯算子

import cv2 as cv

def lapalian_demo(image):
    dst = cv.Laplacian(image, cv.CV_32F)
    lpls = cv.convertScaleAbs(dst)
    cv.imshow("lapalian_demo", lpls)

image = cv.resize(cv.imread("01.jpg"), (300,400))
cv.imshow("image", image)
lapalian_demo(image)
cv.waitKey(0)
cv.destroyAllWindows()

• 自定義拉普拉斯算子

import cv2 as cv
import numpy

def lapalian_demo(image):
    kernel = numpy.array([[0, 1, 0], [1, -4, 1], [0, 1, 0]])
    lpls = cv.filter2D(image, cv.CV_32F, kernel=kernel)
    cv.imshow("lapalian_demo", lpls)

image = cv.resize(cv.imread("01.jpg"), (300,400))
cv.imshow("image", image)
lapalian_demo(image)
cv.waitKey(0)
cv.destroyAllWindows()

2. Canny邊緣提取

2.1 Canny演算法介紹

• Canny是邊緣檢測演算法，是1986年提出的

步驟：對彩色影像進行高斯模糊，去除噪聲，再進行灰度轉換，對影像求取梯度，根據影像角度實作非最大信號壓制，進行高低閾值過濾

1. 高斯模糊：GaussianBlur
2. 灰度轉換：cvtColor
3. 計算梯度：Sobel/Scharr
4. 非最大信號抑制
5. 高低閾值輸出二值影像

2.2 代碼演示

import cv2 as cv
import numpy

def edge_demo(image):
    blurred = cv.GaussianBlur(image, (3, 3), 0)
    gray = cv.cvtColor(blurred, cv.COLOR_BGR2GRAY)
    # x gradient
    gradx = cv.Sobel(gray, cv.CV_16SC1, 1, 0)
    grady = cv.Sobel(gray, cv.CV_16SC1, 0, 1)
    # edge 高閾值為150 低閾值為50
    edge_output = cv.Canny(gradx, grady, 50, 150)
    cv.imshow("Canny Edge", edge_output)

    dst = cv.bitwise_and(image, image, mask=edge_output)
    cv.imshow("Color Edge", dst)

image = cv.resize(cv.imread("01.jpg"), (300,400))
cv.imshow("image", image)
edge_demo(image)
cv.waitKey(0)
cv.destroyAllWindows()

3. 直線檢測

3.1 霍夫直線變換介紹

• Hough Line Transform用來做直線檢測
• 前提條件：邊緣檢測已經完成
• 需要理解：平面空間到極坐標空間轉換

3.2 代碼演示

import cv2 as cv
import numpy

def line_detection(image):
    gray = cv.cvtColor(image, cv.COLOR_BGR2GRAY)
    edges = cv.Canny(gray, 50, 150, apertureSize=3)
    lines = cv.HoughLines(edges, 1, numpy.pi/180, 200)
    for line in lines:
        rho, theta = line[0]
        a = numpy.cos(theta)
        b = numpy.sin(theta)
        x0 = a * rho
        y0 = b * rho
        x1 = int(x0 + 1000 * (-b))
        y1 = int(y0 + 1000 * (a))
        x2 = int(x0 - 1000 * (-b))
        y2 = int(y0 - 1000 * (a))
        cv.line(image, (x1, y1), (x2, y2), (0, 0, 255), 2)
    cv.imshow("image lines", image)


image = cv.resize(cv.imread("star.jpg"), (300,400))
cv.imshow("image", image)
line_detection(image)
cv.waitKey(0)
cv.destroyAllWindows()