Since my postgraduate study, I have been engaged in theoretical and experimental research in computational optical imaging, including non-line-of-sight (NLoS) imaging and single-pixel imaging, and have accumulated extensive research experience in this field. Relevant research is introduced as follows.

(1) Non-line-of-sight imaging

For the reconstruction algorithm of NLoS imaging, we recently drew an analogy to the absorption of surrounding dust by celestial bodies and proposed an absorption backprojection algorithm that mimics the law of universal gravitation. This algorithm absorbs low-confidence voxels in the hidden space into high-confidence voxels, thereby eliminating most artifacts around objects and achieving superior reconstruction results compared to related algorithms. The relevant research results were published in Appl. Phys. Lett. [Non-line-of-sight imaging with absorption backprojection, Appl. Phys. Lett. 123, 054001 (2023)].

Our achievements in NLoS imaging.

The widely used backprojection (BP) algorithm introduces significant artifacts in reconstruction due to spatial position uncertainty, resulting in blurred reconstructions. We recently proposed a BP-based artifact self-cancellation (ACS) algorithm, which constructs a new time-of-flight histogram by removing partial artifact information and performs backprojection to substantially reduce artifacts. We validated our method using our own and openly available datasets, demonstrating that the ASC algorithm enables efficient and stable reconstruction under both confocal and non-confocal conditions, with reconstruction quality comparable to other advanced algorithms. We also employed a Ground truth-independent evaluation metric to assess our reconstruction results and determine parameter values in the algorithm. The relevant research results were published in Optics & Laser Technology [Non-line-of-sight imaging with adaptive artifact cancellation, Optics & Laser Technology, 182 (2024)].

Our achievements in NLoS imaging. The left image shows the algorithm flowchart, and the right image compares the reconstruction effects of our algorithm with existing algorithms.

For transmission-type NLoS scenarios, imaging through strongly scattering media in noisy environments presents significant challenges. To address this issue, we proposed an imaging system based on an upconversion single-photon detector. This upconversion detector exhibits exceptional noise resistance through ultra-narrow optical time gating, quantum mode selectivity, and parametric frequency upconversion techniques. To validate the system's noise resistance, we experimentally compared the imaging quality of near-infrared detectors and upconversion detectors under varying integration times, noise intensities, and object surface types. The experimental results demonstrate that, compared to near-infrared detectors, the upconversion detector exhibits superior robustness in high-noise environments, enabling clear image acquisition through strongly scattering media with 12 optical thickness units and a signal-to-noise ratio as low as 1/400. This work was published in Optics Letters [Upconversion imaging through scattering media in noisy environments, Opt. Lett., 50 (2025)].

Our research achievements in transmission-type NLoS imaging. The left image shows the experimental setup, and the right image compares the reconstruction effects of our upconversion imager with conventional near-infrared imaging methods.

(2) Single-pixel imaging

We have conducted research in single-pixel imaging, incoherent light interference, two-photon entangled double-slit interference at different frequencies, and single-pixel imaging of phase objects. (1) Traditional correlated imaging methods cannot image phase objects due to the use of bucket detection for the detected light intensity. To address this issue, based on previous theoretical research, we developed a control program for synchronous rapid acquisition by three CCD detectors, meeting the system stability requirements of the interferometer and experimentally achieving thermal light correlated imaging of phase objects. The relevant experimental work was published in Appl. Phys. Lett. (2) To overcome the drawbacks of requiring a large number of samples and long acquisition times in correlated imaging, we innovatively proposed utilizing the wavelength resources of the illumination source and randomly modulating light at different wavelengths to significantly reduce the acquisition time for correlated imaging. The relevant experimental work on multi-wavelength correlated imaging was published in Phys. Rev. A. (3) During theoretical research on light field intensity correlation, we discovered significant differences in imaging quality between bucket-detected and point-detected correlated imaging under different illumination source scenarios. To validate this theoretical finding, we designed and completed experiments, and the relevant experimental work comparing the imaging quality of bucket-detected and point-detected correlated imaging was published in Opt. Comm. (4) To enhance the practicality of correlated imaging, we also investigated its application in mobile phone imaging scenarios, with the relevant research published in Opt. Eng. (5) In addition to studying correlation phenomena in classical light, we also explored higher-order correlation phenomena in entangled light fields. For example, higher-order correlations exist between entangled two-photon pairs at different frequencies. The experimental work on two-photon entangled Young's double-slit interference at different frequencies was published in Sci. Rep

Published papers 

[1]H. Zhou, Z. Chen, J. Qiu, S. Zhong, D. Zhang, T. Wang, Q. Liu, and T. Yu, Non-line-of-sight imaging with adaptive artifact cancellation, Optics & Laser Technology 182, 112081 (2025).

[2]H. Le, J. Fang, J. Lin, D. Zhang, T. Wang, Q. Liu, and T. Yu, Upconversion imaging through scattering media in noisy environments, Opt. Lett. (2025).

[3]H. Zhou, D. Zhang, T. Wang, Q. Liao, and T. Yu, Non-line-of-sight imaging with absorption backprojection, Applied Physics Letters 123, 054001 (2023).

[4]J.-T. Liu, Y. Zhang, X. Cai, J. Huang, K. Luo, H. Li, D. Zhang, and Z. Wu, Robust binarized data analysis with computational ghost imaging, Optik 272, 170378 (2023).

[5]J.-T. Liu, J. Li, J.-B. Huang, X.-M. Cai, X. Deng, and D.-J. Zhang, Single-pixel ghost imaging based on mobile phones, Opt. Eng. 58, 1 (2019).

[6]D.-J. Zhang, R. Yin, T.-B. Wang, Q.-H. Liao, H.-G. Li, Q. Liao, and J.-T. Liu, Ghost imaging with bucket detection and point detection, Optics Communications 412, 146 (2018).

[7]D.-J. Zhang, S. Wu, H.-G. Li, H.-B. Wang, J. Xiong, and K. Wang, Young’s double-slit interference with two-color biphotons, Sci Rep 7, 17372 (2017).

[8]D.-J. Zhang, H.-G. Li, Q.-L. Zhao, S. Wang, H.-B. Wang, J. Xiong, and K. Wang, Wavelength-multiplexing ghost imaging, Phys. Rev. A 92, 013823 (2015).

[9]D.-J. Zhang, Q. Tang, T.-F. Wu, H.-C. Qiu, D.-Q. Xu, H.-G. Li, H.-B. Wang, J. Xiong, and K. Wang, Lensless ghost imaging of a phase object with pseudo-thermal light, Applied Physics Letters 104, 121113 (2014)