Abstract: Traditional image segmentation algorithms usually can’t obtain expected effect facing complex images such as container code images with complex background and bad illumination. A novel segmentation algorithm based on valid gradient information is proposed to solve the problem. By statistical analyzing the gradient information of the boundary between the target and the background, the thresholds can be obtained directly and used to segment the image. The experiment results show that the algorithm can get best performance evaluation. Finally, the algorithm has good practicability and can be used directly in different image segmentation fields.

Key words: Image Segmentation, Gradient Information, Threshold

摘要： 本文以集装箱号码图像为例，在分析比较几种流行的图像分割算法后，提出一种新的基于有效梯度信息的图像分割算法，即直接针对梯度图，对目标与背景边界部分的梯度信息进行统计分析，并在此基础上直接得到图像的阈值，完成分割。实验表明，对于类似集装箱号码的复杂图像，本文方法能得到较其他方法更好的分割效果。此外，本文方法有较好的实用性，能广泛应用于多种图像分割领域。

关键词: 图像分割, 梯度信息, 阈值