在夜幕低垂之际，半岛的灯光如梦似幻，成为摄影爱好者的心头好。想要捕捉这些迷人的夜景，灯光摄影技巧是必不可少的。下面，就让我带你走进灯光摄影的世界，一起探索如何拍出令人陶醉的夜景作品。

一、了解夜景摄影的基本原理

1. 光圈、快门、ISO的关系

夜景摄影中，光圈、快门和ISO是三个关键参数。它们之间的关系如下：

• 光圈：光圈越大，进光量越多，画面越亮；光圈越小，进光量越少，画面越暗。
• 快门：快门速度越快，画面越清晰；快门速度越慢，画面越容易模糊。
• ISO：ISO值越高，画面越亮；ISO值越低，画面越清晰。

2. 景深控制

夜景摄影中，景深较浅，因此要特别注意对焦点的选择。一般来说，将焦点对在画面中的主要光源上，如建筑物、桥梁等，可以让画面更有层次感。

二、灯光摄影实用技巧

1. 选择合适的拍摄地点

想要拍摄出迷人的夜景，首先要选择一个合适的拍摄地点。一般来说，城市中心、江边、海边等地方都是不错的选择。

2. 利用三脚架稳定相机

夜景摄影中，由于快门速度较慢，容易造成画面模糊。因此，使用三脚架可以稳定相机，避免画面模糊。

3. 拍摄时间选择

夜晚的灯光摄影，最佳拍摄时间是在月黑风高的夜晚。此时，天空较暗，灯光效果更加明显。

4. 拍摄手法

（1）长时间曝光

长时间曝光可以捕捉到流动的灯光，如车流、河流等，让画面更具动感。

”`python import cv2 import numpy as np

读取视频

cap = cv2.VideoCapture(‘night_capturing_video.mp4’)

创建空白图像

result = np.zeros((720, 1280, 3), dtype=np.uint8)

while cap.isOpened():

ret, frame = cap.read()
if not ret:
    break

# 长时间曝光
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.pyrDown(frame)
cv2.py