ing at the phenomena of abnormal response and dark signal tailing caused by crosstalk of a linear array detector, we analyze the mechanism of crosstalk generation, establish the RC model which can reproduce the crosstalk image signal waveform. On this basis, we further propose a crosstalk image restoration method. This method is based on the crosstalk RC model and the objective is to recover the normal signal response and eliminate the dark signal tailing. The optimization objective function is established. Based on the target response curves of different frequencies, the model parameters are iterated to obtain the corresponding model parameters when the comprehensive effect of restoration is optimal for each image frequency. After the parameters of the crosstalk model are obtained, the corresponding restoration function is calculated, and the image is restored by the operations in the frequency domain. The infrared images of the linear array detector scanning camera acquired in the laboratory are restored and the results show that the proposed method can effectively restore the normal response of different targets under different image frequencies, reduce the effect of tailing dark signals, and improve the image quality.