Liquid crystal lens is an emerging liquid crystal device that can be electrically controlled to modulate the focus, zoom, and depth measurement without mechanical movement, and thus it is widely employed in many fields such as photographic camera, microscopic imaging, and virtual reality. Due to the anisotropy of the liquid crystal material, the liquid crystal lens can only modulate the extraordinary ray, and the ordinary ray is not modulated, which causes reduced image contrast. Optical imaging systems of liquid crystal lens can be retrofitted with polarizing devices to remove the ordinary ray component of the incident light. However, the utilization of polarizers drastically reduces the optical flux and degrades the imaging quality. Additionally, there are three main polarizer-free imaging techniques. The first one is to adopt blue-phase liquid crystals instead of nematic-phase liquid crystals to prepare liquid crystal lenses. However, the blue-phase liquid crystal features small birefringence effect, narrow temperature range, and high voltage, and has not yet reached the practical level. The second is to leverage multilayer liquid crystals instead of single-layer liquid crystals to modulate the two components of incident light. However, the multilayer liquid crystal structure increases the thickness and fabrication cost of the device. The third one is to apply a polarizer-free imaging algorithm and an innovative combination of unsharp masking models to acquire high-quality images.

The non-ideal low-frequency component introduced by the unmodulated extraordinary ray is decreased by utilizing an optical imaging system of liquid crystal lens with a polarizer-free device to acquire one focused image and one unfocused image respectively and perform image processing on the two images. Meanwhile, the unsharp masking model for polarizer-free imaging is proposed to analyze pixel value changes of the images to estimate the percentage of the ordinary ray component, and then the unfocused and focused images are adopted to obtain a high-contrast image.

The optical properties of the liquid crystal lens in the experiment are examined. The experimental results show that the optical focal length is linearly related to the optical aberration of an ideal glass lens. Additionally, the liquid crystal lens is close to the ideal optical aberration of a glass lens with high imaging quality, and can be adopted as a focusing unit in the imaging system (Fig. 7). A polarizer is placed in front of the liquid crystal lens to remove the ordinary ray component in the incident light, and different voltages are applied to the ends of the liquid crystal lens to capture the ISO 12233 chart. The f_MTF50 is to characterize the resolution capability of the system and evaluate the image quality to determine the optimal operating voltage value (Fig. 10). Meanwhile, we adopt f_MTF50 to characterize the system's resolving power, evaluate the image quality, and then determine the optimal operating voltage value (Fig. 10). Focus and non-focus images are captured, the values are set to process the two images, and the unsharp masking model is adopted for edge detection, with observations conducted on whether there is any abnormal "depression" and "bulge". The polarization direction of the incident polarized light is detected and recorded (Table 1). The relationship between the polarization angle and the ideal and actual values is plotted, and the values obtained by experimental measurements are near the ideal value curve, which indicates that the experimental method can obtain the actual values more accurately and process the images (Fig. 14). Meanwhile, we photograph the actual scene and process the image, and the image obtained by the imaging method of polarizer-free liquid crystal lens is clear and natural with high contrast. The value calculated by the unsharp masking model can be applied to the actual measurement (Fig. 16).

We propose a polarizer-free imaging technique of liquid crystal lens based on the unsharp masking model and experimentally verify the feasibility of the technique. The technique combines the unsharp masking model in image processing to estimate the proportion of the ordinary ray component in ambient light by analyzing the changes of the image pixel. Then, this value is employed to process the focused and unfocused images to reduce the non-ideal low-frequency component introduced by non-modulated unusual light and obtain a natural high-quality image. Additionally, values of the ratio of the o-light component are obtained by simulating the ambient light incidence conditions with different polarization directions, and the experimental results are consistent with the theoretical analysis.