Currently, with the widespread application of embedded technology and the continuous improvement of computational power in mobile terminals, the efficient deployment of algorithms on embedded devices, while maintaining high accuracy and minimizing model size, has become a research hotspot. This paper addresses the challenges of deploying the YOLOv8 algorithm on embedded devices and proposes a novel lightweight object detection algorithm focusing on small object detection. We optimize the model through two key strategies, aiming to achieve lightweight deployment and improve the accuracy of small object detection. Firstly, GhostNet is introduced as the backbone network for YOLOv8 in order to achieve lightweight deployment. By using some cost effective operations to generate redundant feature maps, we not only reduce the number of model parameters while ensuring better detection results, but also improve the speed of the model. Secondly, a new multi-scale attention module is designed to enhance the network’s acquisition of crucial information for small targets, which includes a multi-scale fusion attention mechanism and the Soft-NMS algorithm. The multi-scale fusion attention mechanism captures key features of discriminative small targets in the feature map tensor from both spatial and channel dimensions, suppressing non-key information, reducing the impact of complex and unimportant information in the image, enhancing the network model’s learning ability for important features of small targets. The Soft-NMS method improves accuracy by significantly reduces false positives in the detection results. To validate the performance of our proposed method, we conducted validation experiments on the PASCAL VOC dataset and evaluated the model’s generalization ability on the MS COCO dataset. The experiments results demonstrate that our model achieves a significant improvement in small object detection, with a 5.41% increase in detection accuracy compared to the existing YOLOv8. Meanwhile, FLOPs are reduced by 49.62%, and the number of model parameters is reduced by 48.66%. These results fully confirm the effectiveness of our innovative method in achieving both lightweight deployment and significant efficacy in small object detection tasks.

Finger vein recognition has been widely studied due to its advantages, such as high security, convenience, and living body recognition. At present, the performance of the most advanced finger vein recognition methods largely depends on the quality of finger vein images. However, when collecting finger vein images, due to the possible deviation of finger position, ambient lighting and other factors, the quality of the captured images is often relatively low, which directly affects the performance of finger vein recognition. In this study, we proposed a new model for finger vein recognition that combined the vision transformer architecture with the capsule network (ViT-Cap). The model can explore finger vein image information based on global and local attention and selectively focus on the important finger vein feature information. First, we split-finger vein images into patches and then linearly embedded each of the patches. Second, the resulting vector sequence was fed into a transformer encoder to extract the finger vein features. Third, the feature vectors generated by the vision transformer module were fed into the capsule module for further training. We tested the proposed method on four publicly available finger vein databases. Experimental results showed that the average recognition accuracy of the algorithm based on the proposed model was above 96%, which was better than the original vision transformer, capsule network, and other advanced finger vein recognition algorithms. Moreover, the equal error rate (EER) of our model achieved state-of-the-art performance, especially reaching less than 0.3% under the test of FV-USM datasets which proved the effectiveness and reliability of the proposed model in finger vein recognition.

As the second generation of biometric technology, finger vein recognition has become a research hotspot due to its advantages such as high security, and living body recognition. In recent years, the global pandemic has promoted the development of contactless identification. However, the unconstrained finger vein acquisition process will introduce more uneven illumination, finger image deformation, and some other factors that may affect the recognition, so it puts forward higher requirements for the acquisition speed, accuracy and other performance. Considering the universal, obvious, and stable characteristics of the original finger vein imaging, we proposed a new Region Of Interest (ROI) extraction method based on the characteristics of finger vein image, which contains three innovative elements: a horizontal Sobel operator with additional weights; an edge detection method based on finger contour imaging characteristics; a gradient detection operator based on large receptive field. The proposed methods were evaluated and compared with some representative methods by using four different public datasets of finger veins. The experimental results show that, compared with the existing representative methods, our proposed ROI extraction method is $1/10$th of the processing time of the threshold-based methods, and it is similar to the time spent for coarse extraction in the mask-based methods. The ROI extraction results show that the proposed method has better robustness for different quality images. Moreover, the results of recognition matching experiments on different datasets indicate that our method achieves the best Equal Error Rate (EER) of $0.67\%$ without the refinement of feature extraction parameters, and all the EERs are significantly lower than those of the representative methods.

Parkinson’s disease (PD) is a common disease of the elderly. Given the easy accessibility of handwriting samples, many researchers have proposed handwriting-based detection methods for Parkinson’s disease. Extracting more discriminative features from handwriting is an important step. Although many features have been proposed in previous researches, the insight analysis of the combination of handwriting’s kinematic, pressure, and angle dynamic features is lacking. Moreover, most existing feature is incompletely represented, with feature information lost. Therefore, to solve the above problems, a new feature extraction approach for PD detection is proposed using handwriting. First, built on the kinematic, pressure, and angle dynamic features, we propose a moment feature by composed these three types of features, an overall representation of these three types of features information. Then, we proposed a feature extraction method to extract time-frequency-based statistical (TF-ST) features from dynamic handwriting features in terms of their temporal and frequency characteristics. Finally, we proposed an escape Coati Optimization Algorithm (eCOA) for global optimization to enhance classification performance. Self-constructed and public datasets are used to verify the proposed method’s effectiveness respectively. The experimental results showed an accuracy of 97.95% and 98.67%, a sensitivity of 98.15% (average) and 97.78%, a specificity of 99.17% (average) and 100%, and an AUC of 98.66% (average) and 98.89%. The code is available at https://github.com/dreamhcy/MLforPD.