| |
2023/4/3 |
|
| |
We reported
a novel autofocusing method which allows a digital projector to find the
focus position without using a camera. Instead, the method works by
repeatedly projecting three fringe patterns and determining the focus
position rapidly with a photodiode. Inspired by the method, autofocusing
Fourier single-pixel imaging is achieved. To our best knowledge, it is
for the first time that single-pixel imaging with autofocusing is
reported. |
|
| |
|
|
| |
2023/1/1 |
|
| |
We reported a single-pixel imaging method which allows for full-resolution, wide-field-of-view, and high-quality imaging. On the basis of fast Fourier single-pixel imaging, we proposed a novel error diffusion kernel for pattern dithering and binarize the patterns with different dithering strategies. As such, one could obtain two or more single-pixel reconstructions. By synthesizing the reconstructed images, a high-quality image can be obtained. As experimentally demonstrated, two raw reconstructions are sufficient for the final high-quality reconstruction. In addition, the method allows one to make full use of the active area allowed by the spatial light modulator to achieve wide-field-of-view imaging. This work was published in OPTICS LETTERS. |
|
| |
|
|
| |
2022/8/10 |
|
| |
We proposed a light-field microscopy that can achieve full-resolution imaging. By extending the principle of dual photography from real space to Fourier space, the proposed method uses a spatial light modulator placed at the image plane as a virtual 2D detector to record the 2D spatial distribution of the image, and meanwhile, uses a real 2D detector placed at the Fourier plane of the image to record the angles of the light rays. The Fourier-spectrum signals recorded by each pixel of the real 2D detector can be used to reconstruct a perspective image through single-pixel imaging. Based on the perspective images reconstructed by different pixels, we experimentally demonstrated that the camera can digitally refocus on objects at different depths. The camera can achieve light-field imaging with full resolution and provide an extreme depth of field. The method provides a new idea for developing full-resolution light-field cameras. This work was published in PHOTONICS. |
|
| |
|
|
| |
2022/4/15 |
|
| |
广东省科技创新大会在广州召开，会上颁发了2021年度广东省科学技术奖。课题组的《傅里叶单像素成像理论与方法》项目获得了自然科学奖二等奖。 |
|
| |
|
|
| |
2022/4/1 |
|
| |
We proposed a method for illumiantion angle calibration. The key to the method is using a double-sided mask to encode the information about illumiantion angle. The mask is fabricated with a ring array pattern on its top surface and a disk array pattern on its bottom surface. The both patterns are concentric. However, there will be a relative shift between the two patterns, when the mask is under oblique illumination. Thus, we are able to retrieve the illumination angle information by measuring the amount of shift. This work was published in OPTICS LETTERS. |
|
| |
|
|
| |
2022/3/21 |
|
| |
We proposed a imaging-free method for fast-moving objects classification. The method is inspired by single-pixel imaging. The method uses orthogonal transform basis patterns to illuminate the target moving object and uses a single-pixel detector to collect the resulting light signals. By training a artifical neural network, the network is capable of classifying the target moving object from the input single-pixel measurements. This work was published in PHOTONICS. |
|
| |
|
|
| |
2022/2/25 |
|
| |
We proposed a new ringing artifacts elimination method to improve the imaging quality of Fourier single-pixel imaging. Wit the proposed method, Fourier single-pixel imaging is now able to achieve ringing artifacts free imaging for a dynamic scene. This work was published in OPTICS LETTERS. |
|
| |
|
|
| |
2021/12/7 |
|
| |
课题组针对运动物体提出了一种光电混合神经网络，利用空间光调制和单像素探测对物体进行分类。实验证明，该方法可以长时间连续地对快速旋转的手写数字进行识别，其运动物体分类能力超过了人眼视觉。相关论文《基于光电混合神经网络的单像素快速运动物体分类（特邀）》发表在《红外与激光工程》。 |
|
| |
|
|
| |
2021/8/9 |
|
| |
We proposed a new sampling strategy, termed "Gasussian random samping", for efficient Fourier single-pixel imaging. This sampling strategy can effecitvely reduce the number of measurements to achieve fast single-pixel imaging. The key to the strategy is to adopt density-varying sampling in the Fourier domain based on the importance of Fourier coefficients. This work was published in PHOTONICS. |
|
| |
|
|
| |
2021/7/31 |
|
| |
This site launched. |
|
| |
|
|
Contact us