针对传统算法模型先检测、后识别、再定位导致执行效率较差的问题,提出了一种基于协同显著性检测的服务机器人空间物体快速定位方法.利用RGB-D传感器获取N对包含待定位物体的RGB图像与深度图像,将待定位物体看作协同显著性目标,在RGB图像中充分挖掘单幅图像显著性传播机理,构建基于图像间显著性传播和图像内流形排序的两阶段引导协同显著性检测模型,同时排除背景和非协同显著性物体,得到协同显著性物体区域的像素坐标集合.进一步利用RGB图像与深度图像的对应关系确定物体质心的空间坐标,实现对空间物体的快速定位.最后将所提方法在iCoseg标准数据库和经手眼标定后的服务机器人机械臂抓取平台上进行实验.实验结果表明,该方法一致优于现有的5种协同显著性检测算法,且满足抓取系统的实时性需求,在复杂背景、多目标干扰以及光照变化时具有较高的准确性和鲁棒性. In order to solve the problem of poor efficiency in the first detection, post-recognition and re-positioning of the traditional algorithm model, a fast locating method of service robot based on collaborative saliency detection is proposed. By using RGB-D sensor, RGB image and depth image, the object to be positioned is regarded as a coherence saliency target, the saliency transmission mechanism of single image is fully tapped in the RGB image, and a two-phase guiding cooperation based on salient propagation and intra-image manifold ranking is constructed Significant detection model and exclude the background and non-cooperative saliency objects to get the pixel coordinate set of the region of synergistic saliency.Further use of the correspondence between RGB image and depth image to determine the space coordinates of the object centroid to achieve the rapid Finally, the proposed method is tested in the iCoseg standard database and the service robotic arm grasping platform with eyes calibrated.Experimental results show that the proposed method is superior to the existing five kinds of cooperative saliency detection algorithms, and satisfies Catch the system’s real-time needs, in complex background, multi-target interference and lighting changes with High accuracy and robustness.