基于图像增强的温室图像特征点提取

针对温室环境中，由于相机拍摄图像噪声大，光照不均匀，对比度较低等会导致ORB特征点较难提取或者提取较少，从而影响匹配效果的问题，提出了一种能够减少图像噪声，增强图像对比度的ORB特征点提取算法。首先，对温室图像进行高斯滤波去除图像噪声，然后使用对比度受限的自适应直方图增强图像对比度，增加ORB特征点提取数量。经过实验表明，提出的改进ORB算法对温室中ORB特征点的提取和匹配效果有较大提升。

Feature Points Extraction of Greenhouse Image Based on Image Enhancement

Aiming at the problem that ORB feature points are difficult to be extracted or less to be extracted due to the high noise, uneven illumination and low contrast of images taken by camera in greenhouse environment, thus affecting the matching effect, an ORB feature points extraction algorithm was proposed to reduce image noise and enhance image contrast at the same time. Firstly, Gaussian filtering is applied to the greenhouse image to remove the image noise, and then adaptive histogram with limited contrast is used to enhance the image contrast and increase the number of ORB feature points extracted. The experimental results show that the proposed improved ORB algorithm can greatly improve the extraction and matching effect of ORB feature points in greenhouse.