Aiming at the problems that the existing Deep Convolutional Neural Network (DCNN) have insufficient defect image feature extraction ability, few recognition types and long reasoning time in hydraulic tunnel defect recognition tasks, an autonomous defect recognition method based on dynamic feature distillation was proposed. Firstly, the deep curve estimation network was used to optimize the image to improve the image quality in low illumination environment. Secondly, the dynamic convolution module with attention mechanism was constructed to replace the traditional static convolution, and the obtained dynamic features were used to train the teacher network to obtain better model feature extraction ability. Finally, a dynamic feature distillation loss was constructed by fusing the discriminator structure in the knowledge distillation framework, and the dynamic feature knowledge was transferred from the teacher network to the student network through the discriminator, so as to achieve the high-precision recognition of six types of defects while significantly reducing the model reasoning time. In the experiments, the proposed method was compared with the original residual network on a hydraulic tunnel defect dataset of a hydropower station in Sichuan Province. The results show that this method has the recognition accuracy reached 96.15%, and the model parameter amount and reasoning time reduced to 1/2 and 1/6 of the original ones respectively. It can be seen from the experimental results that fusing the dynamic feature distillation information of the defect image into the recognition network can improve the efficiency of hydraulic tunnel defect recognition.

The distributed state of noise existing among different bands of hyperspectral remote sensing image is complex, so the traditional denoising methods are hard to achieve the desired effect. In order to solve this problem, based on Principal Component Analysis (PCA), a novel denoising method for hyperspectral data was proposed combining with noise estimation and dictionary learning. Firstly, a group of the principal component images were achieved from the original hyperspectral data by using the PCA transform, which were divided into clear image group and noisy image group according to the corresponding energy. Then, according to any band image from noisy hyperspectral data, the noise standard deviation of the image was estimated via a noise estimation method based on Singular Value Decomposition (SVD). Meanwhile, combining this noise estimation method with denoising method via K-SVD dictionary learning, a new dictionary learning denoising method with adaptive noise estimation characteristics was proposed and applied to denoise those images from noisy image group with low energy where noise mainly existed. Finally, the final denoising image was obtained by weighted fusion according to the corresponding energy of each principal component image. The experimental results on simulated and real hyperspectral remote sensing data show that, compared with PCA, PCA-Bish and PCA-Contourlet, the Peak Signal-to-Noise Ratio (PSNR) of the image denoised by the proposed algorithm is improved by 1-3 dB, and more detailed information and better visual effect of the denoised image by the proposed method are achieved.