Without increasing the CCD pixels,the super-resolution reconstruct ion of sub-pixel imaging by increasing the temporal and spatial sampling frequencies in CCDs is o ne of the approaches to improve the system resolution.However,owing to the increase of sampling freque ncy,there exists aliasing in data collected by CCDs and blurring in a reconstructed high-resolut ion image,as a result of which the resolution is far away from the ideal value.To resolve the problem,a n algorithm of super-resolution reconstruction,based on sub-pixel imaging achieved by multip le linear array CCDs,is put forward.First of all,an interpolation model on high-resolution grid is es tablished.Next,blur kernels in an image with high-resolution are identified in linear array and scanning direc tion,respectively to obtain the blur kernel in a frame.Last,the gradient smooth Richard-Lucy algorithm wi th Neumman boundary conditions is employed to deblur the image and the ringing effects are inhibited as well.Experimental results prove that the gray mean grade (GMG) of super-resolution images reconstructed with the al gorithm compared with bilinear interpolation is improved by 7.63,which outperforms in both vis ual q uality and details. Super-resolution reconstruction of sub-pixel imaging in mu ltiple linear array CCDs can be achieved by this algorithm to obtain higher syst em resolution.