本文提出了一种在复杂场景中实时检测、识别多个目标物体的方法。利用由矩阵码(matrixcode)构成的2D矩阵来识别复杂场景中多个目标物。算法不仅能适应目标物体在旋转、缩放、平移变换下的不变性识别,而且能适应仿射及射影变换下的不变性识别。试验结果证明所提供的算法在识别和监视系统中有很好的鲁棒性和实时性。 In this paper, a method of real-time detection and recognition of multiple target objects in complex scenes is proposed. Multiple matrices in complex scenes are identified using 2D matrices composed of matrix codes. The algorithm can not only adapt to the invariance identification of the target object under rotation, scaling and translation transformation, but also adapt to the invariance identification under the affine and projective transformations. The experimental results show that the proposed algorithm has good robustness and real-time performance in the identification and monitoring system.