为降低对比度对网格定位和基因点分割的影响，提出了一种基因芯片图像的自适应对比度增强方法。基于四阶矩进行图像对比度计算，并统计图像背景灰度，从而实现仅针对基因点的自适应对比度增强。通过6个不同数据集的79幅图像实验表明：该对比度增强方法无需人工干预及参数输入，可自动增强图像对比度，鲁棒性强；对比度增强后的图像网格定位准确率普遍提高，最高可达25％。%An automatic algorithm for enhanced contrast is proposed for microarray imaging to reduce the low precision of gridding and segmentation caused with low contrast images. Contrast was first computed based on the fourth order moment and a background gray value estimated. Contrast was thus enhanced on the spot only. Experiments on 79 images drawn from six different data sets were then done. The results show that the proposed method can enhance the image contrast with good robustness without any parameter input or other human intervention. In addition the gridding accuracy is increased with enhanced contrast up to a maximum of 25%.
展开▼
