python图片裁剪&矩形复原
Posted 怪皮蛇皮怪
tags:
篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了python图片裁剪&矩形复原相关的知识，希望对你有一定的参考价值。
图片裁剪&矩形复原
这篇博文的目的是从图片中提取一定的矩形区域作为新的图片
简单来说：我给我的房间拍了张照片，其中我的桌子上有一个相框，相框里有全家福
现在我的全家福丢了，所以我想从以前的房间照片里，提取出其中的全家福并重新打印一张（忽视画质问题）
现在我就是这么个目的
实现效果
该方法通过鼠标选择四个点，在手动设置生成的矩形长宽后生成变换后的矩形
这篇需要设置矩形长宽，后续的2.0版本不需要设置
操作流程
首先复制代码会吧？
1.有张照片
这是网上随便找的一张照片，自行保存测试
2.看看照片
运行代码，其中show_img函数是展示照片
3.选择角点
按照左上，右上，右下，左下的顺序选择四个角点
如果担心自己选不好，可以直接去除我代码里的points的注释，那是我自己用的原版
4.最终结果
代码分析
import 没什么好说的
#如果python没有安装cv2，那么就安装python-opencv就好
import cv2 as cv
import numpy as np
获取图片的长宽
#输入cv.imread后的图片，通过点击四个点选择要裁剪的部分
def get_window_size(src, bound=600):
h,w = src.shape[0], src.shape[1]
if h > w:
h, w = bound, int(w*bound/h)
else:
h, w = int(h*bound/w), bound
return (h, w)
通过鼠标获取图片的坐标点，顺序是左上，右上，右下，左下
class Indexer:
def __init__(self, bound=4):
self.id = 0
self.bound = bound
def get_id(self):
self.id = (self.id + 1)
return (self.id)
def on_EVENT_LBUTTONDOWN(event, x, y, flags, param):
if event == cv.EVENT_LBUTTONDOWN:
img = param['src']
win_name = param['window']
indexer = param['indexer']
points = param['points']
curr_id = indexer.get_id()
points.append((x, y))
print('第{}个顶点: ({},{})'.format(curr_id, x, y))
cv.circle(img, (x, y), 10, (0, 0, 255), thickness=2)
cv.putText(
img,
str(curr_id), # 文字
(x, y), # 坐标
cv.FONT_HERSHEY_PLAIN,
5, # 字号
(0, 0, 255), # 字体颜色
thickness=2 # 粗细
)
cv.imshow(win_name, img)
#输入cv.imread后的图片，通过点击四个点选择要裁剪的部分
def get_points(src):
points = []
indexer = Indexer()
h, w=get_window_size(src)
win_name = 'get_points'
cv.namedWindow(win_name, cv.WINDOW_NORMAL)
cv.resizeWindow(win_name, width=w, height=h)
cv.imshow(win_name, src)
cv.setMouseCallback(win_name, on_EVENT_LBUTTONDOWN,
param={'src': src, 'window': win_name, 'indexer': indexer, 'points': points})
cv.waitKey(0)
cv.destroyAllWindows()
if len(points)>4:
return points[0:4]
# print(points)
# points=[(2, 14), (90, 50), (87, 194), (1, 204)]
return points
#输入cv.imread后的图片，展示图片长什么样
def show_img(src):
win_name = 'show_img'
h, w=get_window_size(src)
cv.namedWindow(win_name, cv.WINDOW_NORMAL)
cv.resizeWindow(win_name, width=w, height=h)
cv.imshow(win_name, src)
cv.waitKey(0)
cv.destroyAllWindows()
将图片截取，并按照指定的长宽比恢复成矩形
def photo_cut_restore(src,points,H,W):
target_points = [(0, 0), (W, 0), (W, H), (0, H)]
points, target_points = np.array(points, dtype=np.float32), np.array(target_points, dtype=np.float32)
M = cv.getPerspectiveTransform(points, target_points)
# print('透视变换矩阵:', M)
result = cv.warpPerspective(src_copy, M, (0, 0))
result = result[:H, :W]
win_name = 'Result'
cv.namedWindow(win_name, cv.WINDOW_NORMAL)
cv.resizeWindow(win_name, width=W, height=H)
cv.imshow(win_name,result)
cv.waitKey(0)
cv.destroyAllWindows()
return result
主程序
if __name__ == '__main__':
path = './1.jpg'
src = cv.imread(path)
src_copy = src.copy()
show_img(src)
W = 20
H = 20
# points=[(112, 308), (175, 310), (176, 369), (113, 369)]
points=get_points(src)
n = 20
W = int(W * n)
H = int(H * n)
result=photo_cut_restore(src_copy,points,H,W)
output_file = 'result.jpg'
cv.imwrite(output_file, result)
懒人一键复制代码
诶，气不气，好不容易一段段复制完，结果最后居然有一键复制的地方
import cv2 as cv
import numpy as np
#输入cv.imread后的图片，通过点击四个点选择要裁剪的部分
def get_window_size(src, bound=600):
h,w = src.shape[0], src.shape[1]
if h > w:
h, w = bound, int(w*bound/h)
else:
h, w = int(h*bound/w), bound
return (h, w)
class Indexer:
def __init__(self):
self.id = 0
def get_id(self):
self.id = (self.id + 1)
return (self.id)
def on_EVENT_LBUTTONDOWN(event, x, y, flags, param):
if event == cv.EVENT_LBUTTONDOWN:
img = param['src']
win_name = param['window']
indexer = param['indexer']
points = param['points']
curr_id = indexer.get_id()
points.append((x, y))
print('第{}个顶点: ({},{})'.format(curr_id, x, y))
cv.circle(img, (x, y), 10, (0, 0, 255), thickness=2)
cv.putText(
img,
str(curr_id), # 文字
(x, y), # 坐标
cv.FONT_HERSHEY_PLAIN,
5, # 字号
(0, 0, 255), # 字体颜色
thickness=2 # 粗细
)
cv.imshow(win_name, img)
#输入cv.imread后的图片，通过点击四个点选择要裁剪的部分
def get_points(src):
points = []
indexer = Indexer()
h, w=get_window_size(src)
win_name = 'get_points'
cv.namedWindow(win_name, cv.WINDOW_NORMAL)
cv.resizeWindow(win_name, width=w, height=h)
cv.imshow(win_name, src)
cv.setMouseCallback(win_name, on_EVENT_LBUTTONDOWN,
param={'src': src, 'window': win_name, 'indexer': indexer, 'points': points})
cv.waitKey(0)
cv.destroyAllWindows()
if len(points)>4:
return points[0:4]
# print(points)
# points=[(2, 14), (90, 50), (87, 194), (1, 204)]
return points
#输入cv.imread后的图片，展示图片长什么样
def show_img(src):
win_name = 'show_img'
h, w=get_window_size(src)
cv.namedWindow(win_name, cv.WINDOW_NORMAL)
cv.resizeWindow(win_name, width=w, height=h)
cv.imshow(win_name, src)
cv.waitKey(0)
cv.destroyAllWindows()
def photo_cut_restore(src,points,H,W):
target_points = [(0, 0), (W, 0), (W, H), (0, H)]
points, target_points = np.array(points, dtype=np.float32), np.array(target_points, dtype=np.float32)
M = cv.getPerspectiveTransform(points, target_points)
# print('透视变换矩阵:', M)
result = cv.warpPerspective(src_copy, M, (0, 0))
result = result[:H, :W]
win_name = 'Result'
cv.namedWindow(win_name, cv.WINDOW_NORMAL)
cv.resizeWindow(win_name, width=W, height=H)
cv.imshow(win_name,result)
cv.waitKey(0)
cv.destroyAllWindows()
return result
if __name__ == '__main__':
path = './3.jpg'
src = cv.imread(path)
src_copy = src.copy()
# show_img(src)
W = 20
H = 20
# points=[(124, 182), (181, 177), (180, 243), (125, 266)]
points=get_points(src)
print(points)
n = 20
W = int(W * n)
H = int(H * n)
result=photo_cut_restore(src_copy,points,H,W)
output_file = 'result.jpg'
cv.imwrite(output_file, result)
以上是关于python图片裁剪&矩形复原的主要内容，如果未能解决你的问题，请参考以下文章