基于MS-RG混合图像分割模型的道路检测研究

道路场景因其结构的多样性、纹理变化的复杂性和自然曝光的不稳定性，使得传统基于道路分割的道路检测方法大多存在信息冗余，并且存在边界丢失、模糊等质量问题.本文首先在道路图像上使用 Meanshift均值漂移算法，通过空间内的概率密度呈梯形上升去寻找局部最优，并搜索属于同一模点的像素然后生成获得超像素块.然后利用 Meanshift算法获得的聚类超像素块进行多种子点区域生长，规范生长规则，克服不能得到封闭边界的缺陷，改进道路图像的分割效果.实验结果表明，本文提出的模型适用性强，相比于传统方法有效地提升了分割准确性和实时性，可准确识别出图像中的道路信息，确保车辆能够行驶在可行驶区域上.

Road Detection Based on MS-RG Hybrid Image Segmentation Model

Due to the diversity of its structure, the complexity of texture changes and the instability of natural exposure, road scenes based on road segmentation mostly have information redundancy, and there are quality problems such as boundary loss and blur. In this paper, we first used the Meanshift algorithm on the road image to find the local optimum by the trapezoidal rise of the probability density in the space, and search for the pixels with the same modulus and then get together to form the super pixel block. Then, the clustering super- pixel block obtained by the Meanshift algorithm was used to perform a variety of sub-point region growth, standardized the growth rule, overcome the defect that the closed boundary cannot be obtained, improve the segmentation effect of the road image. The experimental results show that the proposed method has strong applicability, and it can effectively improve the segmentation accuracy and real-time performance compared with the traditional method, and can accurately identify the road information in the image to ensure that the vehicle can travel on the travelable area.