论文部分内容阅读
空间邻近目标的存在对红外传感器的信号处理提出了超分辨的新要求。通过对红外焦平面的目标成像进行建模,推导了基于最小二乘准则的超分辨目标函数,针对传统最优化方法,对起始估计位置要求高、对高维目标函数计算复杂的缺点,引入粒子群优化算法,优化超分辨目标函数,联合估计出目标在红外焦平面的投影位置和辐射强度,实现对空间邻近目标的红外超分辨。结果表明,在模型最小二乘准则下,基于粒子群优化的超分辨算法性能优于传统的最陡下降法,具备更强的超分辨能力。
The existence of spatially adjacent objects presents a new requirement for super-resolution of the signal processing of infrared sensors. By modeling the target imaging of infrared focal plane, the super-resolution objective function based on least squares criterion is deduced. According to the traditional optimization method, the initial estimation position is high and the calculation of high-dimensional objective function is complicated. Particle swarm optimization (PSO) algorithm and optimization of super-resolution objective function are combined to estimate the projection position and radiation intensity of the target in the infrared focal plane to achieve infrared super-resolution of the target in space. The results show that the performance of the super-resolution algorithm based on PSO is better than the traditional steepest descent method under the model least-squares criterion, and it has stronger super-resolution capability.