Compared with traditional methods, the action recognition model based on 3D convolutional deep neural network captures spatio-temporal features more accurately, resulting in higher accuracy. However, the large number of parameters and computational requirements of 3D models make it difficult to deploy on mobile devices with limited computing power. In order to achieve an efficient video action recognition model, we have analyzed and compared classic lightweight network principles and proposed the 3D-ShuffleViT network. By deeply integrating the self-attention mechanism with convolution, we have introduced an efficient ACISA module that further enhances the performance of our proposed model. This has resulted in exceptional performance in both context-sensitive and context-independent action recognition, while reducing deployment costs. It is worth noting that our 3D-ShuffleViT network, with a computational cost of only 6% of that of SlowFast-ResNet101, achieved 98% of the latter’s Top1 accuracy on the EgoGesture dataset. Furthermore, on the same CPU (Intel i5-8300H), its speed was 2.5 times that of the latter. In addition, when we deployed our model on edge devices, our proposed network achieved the best balance between accuracy and speed among lightweight networks of the same order.

Wireless capsule endoscopy (WCE) has become an important tool for gastrointestinal examination due to its non-invasive nature and minimal patient discomfort. However, the quality of WCE images is often limited by built-in lighting and the complex gastrointestinal environment, particularly in the region filled with small intestinal villi. Additionally, the morphology of these villi usually serves as a crucial indicator for related diseases. To address this, we propose a novel method to enhance the clarity of small intestinal villi in WCE images. Our method uses a guided filter to separate the low- and high-frequency components of WCE images. Illumination gain factors are calculated from the low-frequency components, while gradient gain factors are derived from Laplacian convolutions on different regions. These factors enhance the high-frequency components, combined with the original image. This approach improves edge detail while suppressing noise and avoiding edge overshoot, providing clearer images for diagnosis. Experimental results show that our proposed method achieved a 45.47% increase in PSNR compared to classical enhancement algorithms, a 12.63% improvement in IRMLE relative to the original images, and a 31.84% reduction in NIQE with respect to the original images.

Feature matching is crucial in image recognition. However, blurring caused by illumination changes often leads to deviations in local appearance-based similarity, resulting in ambiguous or false matches—an enduring challenge in computer vision. To address this issue, this paper proposes a method named MultS-ORB (Multistage Oriented FAST and Rotated BRIEF). The proposed method preserves all the advantages of the traditional ORB algorithm while significantly improving feature matching accuracy under illumination-induced blurring. Specifically, it first generates initial feature matching pairs using KNN (K-Nearest Neighbors) based on descriptor similarity in the Hamming space. Then, by introducing a local motion smoothness constraint, GMS (Grid-Based Motion Statistics) is applied to filter and optimize the matches, effectively reducing the interference caused by blurring. Afterward, the PROSAC (Progressive Sampling Consensus) algorithm is employed to further eliminate false correspondences resulting from illumination changes. This multistage strategy yields more accurate and reliable feature matches. Experimental results demonstrate that for blurred images affected by illumination changes, the proposed method improves matching accuracy by an average of 75%, reduces average error by 33.06%, and decreases RMSE (Root Mean Square Error) by 35.86% compared to the traditional ORB algorithm.

The images captured by Wireless Capsule Endoscopy (WCE) always exhibit specular reflections, and removing highlights while preserving the color and texture in the region remains a challenge. To address this issue, this paper proposes a highlight removal method for capsule endoscopy images. Firstly, the confidence and feature terms of the highlight region's edges are computed, where confidence is obtained by the ratio of known pixels in the RGB space's R channel to the B channel within a window centered on the highlight region's edge pixel, and feature terms are acquired by multiplying the gradient vector of the highlight region's edge pixel with the iso-intensity line. Subsequently, the confidence and feature terms are assigned different weights and summed to obtain the priority of all highlight region's edge pixels, and the pixel with the highest priority is identified. Then, the variance of the highlight region's edge pixels is used to adjust the size of the sample block window, and the best-matching block is searched in the known region based on the RGB color similarity and distance between the sample block and the window centered on the pixel with the highest priority. Finally, the pixels in the best-matching block are copied to the highest priority highlight removal region to achieve the goal of removing the highlight region. Experimental results demonstrate that the proposed method effectively removes highlights from WCE images, with a lower coefficient of variation in the highlight removal region compared to the Crinimisi algorithm and DeepGin method. Additionally, the color and texture in the highlight removal region are similar to those in the surrounding areas, and the texture is continuous.

To address the issue of increased triangulation uncertainty caused by selecting views with small camera baselines in Structure from Motion (SFM) view selection, this paper proposes a robust error-resistant view selection method. The method utilizes a triangulation-based computation to obtain an error-resistant model, which is then used to construct an error-resistant matrix. The sorting results of each row in the error-resistant matrix determine the candidate view set for each view. By traversing the candidate view sets of all views and completing the missing views based on the error-resistant matrix, the integrity of 3D reconstruction is ensured. Experimental comparisons between this method and the exhaustive method with the highest accuracy in the COLMAP program are conducted in terms of average reprojection error and absolute trajectory error in the reconstruction results. The proposed method demonstrates an average reduction of 29.40% in reprojection error accuracy and 5.07% in absolute trajectory error on the TUM dataset and DTU dataset.

To address the issue of feature descriptors being ineffective in representing grayscale feature information when images undergo high affine transformations, leading to a rapid decline in feature matching accuracy, this paper proposes a region feature descriptor based on simulating affine transformations using classification. The proposed method initially categorizes images with different affine degrees to simulate affine transformations and generate a new set of images. Subsequently, it calculates neighborhood information for feature points on this new image set. Finally, the descriptor is generated by combining the grayscale histogram of the maximum stable extremal region to which the feature point belongs and the normalized position relative to the grayscale centroid of the feature point's region. Experimental results, comparing feature matching metrics under affine transformation scenarios, demonstrate that the proposed descriptor exhibits higher precision and robustness compared to existing classical descriptors. Additionally, it shows robustness when integrated with other descriptors.