We propose a new GPU based technique to render realistic rain in real-time. It consists of two parts: off-line image analysis of rain videos, and real-time particle-based synthesis of rain. Videos of real rain are analyzed to extract the rain mattes; random samples of these rain stroke mattes are then used for online synthesis. We incorporate pre-computed radiance transfer (PRT) in our particle system to take into account the scene radiance. The transfer function of a raindrop can be highly efficiently evaluated with a closed-form solution. Our approach achieves high realism with low computation cost and a small memory footprint. Our technique applies to a variety of scenarios, from synthetic 3D scenes to real videos

Tunable diode laser absorption spectroscopy (TDLAS) is a significant technique for measuring gas concentration in the combustion field. When the gas concentration is detected by wavelength modulation spectroscopy, the second harmonic (2f) signal demodulated by a lock-in amplifier can be analyzed to obtain the gas concentration information. However, the 2f signal will be affected by the white Gaussian noise of electronic equipment and the optical fringe of the standard instrument, which will lead to the reduction of the gas detection accuracy. To solve the above problems, this paper proposes a 2f signal noise reduction algorithm based on the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), detrended fluctuation analysis (DFA), and wavelet soft threshold (WST). Taking CO₂ gas as an example, the 2f signal extracted from the experiment was denoised, and the amplitude of the 2f signal was linearly fitted to the gas concentration. The R² value was 0.9979, and the SNR was 31.9750 dB. The denoising effect is obvious, and the denoising algorithm can better retain the peak information of the second harmonic signal. In this paper, the existing classical noise reduction algorithm is simulated and analyzed. To display the noise reduction effect visually in the time-frequency domain, the Hilbert-Huang transform three-dimensional spectrum is introduced to analyze the spectrum characteristics of the noise reduction signal.

Soft shadows of environmental lighting provide important visual cues in realistic rendering. However, rendering of soft shadows of environmental lighting in real-time is difficult because evaluating the visibility function is challenging. In this work, we present a method to render soft shadows of environmental lighting at real-time frame rates based on hardware-accelerated ray tracing. We assume that the scene contains both static and dynamic objects. To composite the soft shadows cast by dynamic objects with the precomputed lighting of static objects, the incident irradiance occluded by dynamic objects, which is obtained by accumulating the occluded incident radiances over the hemisphere using ray tracing, is subtracted from the precomputed incident irradiance. Conical ray culling is proposed to exclude the rays that cannot intersect dynamic objects, which significantly improves rendering efficiency. Rendering results demonstrate that our proposed method can achieve real-time rendering of soft shadows of environmental lighting cast by dynamic objects.

There are six possible solutions for the surface normal vectors obtained from polarization information during 3D reconstruction. To resolve the ambiguity of surface normal vectors, scholars have introduced additional information, such as shading information. However, this makes the 3D reconstruction task too burdensome. Therefore, in order to make the 3D reconstruction more generally applicable, this paper proposes a complete framework to reconstruct the surface of an object using only polarized images. To solve the ambiguity problem of surface normal vectors, a jump-compensated U-shaped generative adversarial network (RU-Gan) based on jump compensation is designed for fusing six surface normal vectors. Among them, jump compensation is proposed in the encoder and decoder parts, and the content loss function is reconstructed, among other approaches. For the problem that the reflective region of the original image will cause the estimated normal vector to deviate from the true normal vector, a specular reflection model is proposed to optimize the dataset, thus reducing the reflective region. Experiments show that the estimated normal vector obtained in this paper improves the accuracy by about 20° compared with the previous conventional work, and improves the accuracy by about 1.5° compared with the recent neural network model, which means the neural network model proposed in this paper is more suitable for the normal vector estimation task. Furthermore, the object surface reconstruction framework proposed in this paper has the characteristics of simple implementation conditions and high accuracy of reconstructed texture.

Registration of simultaneous polarization images is the premise of subsequent image fusion operations. However, in the process of shooting all-weather, the polarized camera exposure time need to be kept unchanged, sometimes polarization images under low illumination conditions due to too dark result in SURF algorithm can not extract feature points, thus unable to complete the registration, therefore this paper proposes an improved SURF algorithm. Firstly, the luminance operator is used to improve overall brightness of low illumination image, and then create integral image, using Hession matrix to extract the points of interest to get the main direction of characteristic points, calculate Haar wavelet response in X and Y directions to get the SURF descriptor information, then use the RANSAC function to make precise matching, the function can eliminate wrong matching points and improve accuracy rate. And finally resume the brightness of the polarized image after registration, the effect of the polarized image is not affected. Results show that the improved SURF algorithm can be applied well under low illumination conditions.