在目标成像过程中,受红外光学成像系统衍射、大气扰动等因素的影响,目标能量弥散,图像边缘模糊。传统的基于阈值分割的方法获取的几何成像区域目标像素数误差大,制约了小目标辐射强度的测量精度。针对在像平面投影形状为矩形的目标,提出一种几何成像区域像素数测量方法。首先建立目标成像模型,并将成像模型简化至水平和垂直两个方向;然后分别构建目标尺度特征鉴别曲线;最后从特征曲线中提取目标边长几何成像像素数,从而得到了几何成像区域目标像素数。仿真和外场实验验证了所提方法的有效性。

In an imaging process, a target is affected by factors such as optical diffraction of infrared optical imaging system and atmospheric disturbance, resulting in the diffusion of the target energy and blurring of the edges of target images. Conventional methods based on the threshold segmentation algorithm yield considerable errors when measuring the number of pixels of targets in a geometric imaging area. Thus, the measurement accuracy of the radiation intensity of small targets is restricted. This study proposes a method of pixel number measurement in the geometric imaging area for such targets projected as a rectangular shape on an image plane. First, an target imaging model is established, and the model is simplified toward the horizontal and vertical directions. Then, scale feature identification curves of targets are constructed. Finally, the number of geometric imaging pixels of side length in targets is extracted based on identification curves, and the number of pixels of targets in geometric imaging area is calculated. Simulation and experimental results verify the effectiveness of the proposed method.