基于OpenCV实现物理棋盘棋子位置转矩阵的技术问询

需求目标 #

• 通过摄像头拍摄的照片，提取物理棋盘上的棋子位置
• 将位置转换为python-chess库文档中的矩阵格式，用于后续验证棋子移动

前期尝试与问题 #

HoughLines提取网格失败 #

• 尝试用cv2.HoughLines提取棋盘网格但未成功，推测原因是棋盘本身或摄像头鱼眼畸变
• 即使解决网格提取问题，仍不清楚如何获取棋子坐标、区分空方格与有棋子的方格

角点检测的困境 #

• 改用颜色检测思路，通过cv2.goodFeaturesToTrack成功检测到棋盘所有角点，但角点坐标是无序的
• 尝试通过四个外部分角点划分距离获取方格中点，但因摄像头非正上方拍摄存在透视变形，方格大小不一，该方法不可靠
• 当前核心需求：找到属于同一方格的四个角点，进而计算每个方格的中点

初始HoughLines尝试代码 #

import cv2
import numpy as np


def main():
    im = cv2.imread('opencv_frame_0.png')
    #im = cv2.resize(im, (640, 480))  
    #edge = cv2.imread('edge.png', 0)
    edge = cv2.Canny(im, 60, 160, apertureSize=3)
    lines = cv2.HoughLines(edge, 1, np.pi/180, 100, 100, 50)
    for rho, theta in lines[0]:
        a = np.cos(theta)
        b = np.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))
        cv2.line(im, (x1, y1), (x2, y2), (0, 0, 255), 2)
        # TODO: filter the lines by color and line distance

    cv2.imshow('image', im)
    cv2.imshow('edge', edge)
    cv2.waitKey(0)
    cv2.destroyAllWindows()

if __name__ == "__main__":
    main()

角点检测代码 #

import cv2
import numpy as np

def main():
    frame = cv2.imread("chessboard.jpg")

    cv2.namedWindow("Frame", cv2.WINDOW_NORMAL)

    corners = []

    corners = find_chessboard(frame)

    print(corners)

    cv2.waitKey(0)
    cv2.destroyAllWindows()
    return True

def find_chessboard(frame):
    img = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

    corners = cv2.goodFeaturesToTrack(img, 81, 0.01, 10)
    corners = np.int0(corners)

    corners_coordinates = [[]]*len(corners)

    i = 0
    for corner in corners:
        x, y = corner.ravel()
        corners_coordinates[i] = [x, y]
        i = i+1

        cv2.circle(frame, (x, y), 5, (255, 0, 0), -1)
    cv2.imshow("Frame", frame)

    return corners_coordinates

if __name__ is "__main__":

    main()

内容的提问来源于stack exchange，提问作者SmartlessName