Small-object detection is a basic and challenging problem in computer vision tasks. It is widely used in pedestrian detection, traffic sign detection, and other fields. This paper proposes a deep learning small-object detection method based on image super-resolution to improve the speed and accuracy of small-object detection. First, we add a feature texture transfer (FTT) module at the input end to improve the image resolution at this end as well as to remove the noise in the image. Then, in the backbone network, using the Darknet53 framework, we use dense blocks to replace residual blocks to reduce the number of network structure parameters to avoid unnecessary calculations. Then, to make full use of the features of small targets in the image, the neck uses a combination of SPPnet and PANnet to complete this part of the multi-scale feature fusion work. Finally, the problem of image background and foreground imbalance is solved by adding the foreground and background balance loss function to the YOLOv4 loss function part. The results of the experiment conducted using our self-built dataset show that the proposed method has higher accuracy and speed compared with the currently available small-target detection methods.

Given that facial features contain a wide range of identification information and cannot be completely represented by a single feature, the fusion of multiple features is particularly significant for achieving a robust face recognition performance, especially when there is a big difference between the test sets and the training sets. This has been proven in both traditional and deep learning approaches. In this work, we proposed a novel method named C2D-CNN (color 2-dimensional principal component analysis (2DPCA)-convolutional neural network). C2D-CNN combines the features learnt from the original pixels with the image representation learnt by CNN, and then makes decision-level fusion, which can significantly improve the performance of face recognition. Furthermore, a new CNN model is proposed: firstly, we introduce a normalization layer in CNN to speed up the network convergence and shorten the training time. Secondly, the layered activation function is introduced to make the activation function adaptive to the normalized data. Finally, probabilistic max-pooling is applied so that the feature information is preserved to the maximum extent while maintaining feature invariance. Experimental results show that compared with the state-of-the-art method, our method shows better performance and solves low recognition accuracy caused by the difference between test and training datasets.

In order to solve the problem of face recognition in complex environments being vulnerable to illumination change, object rotation, occlusion, and so on, which leads to the imprecision of target position, a face recognition algorithm with multi-feature fusion is proposed. This study presents a new robust face-matching method named SR-CNN, combining the rotation-invariant texture feature (RITF) vector, the scale-invariant feature transform (SIFT) vector, and the convolution neural network (CNN). Furthermore, a graphics processing unit (GPU) is used to parallelize the model for an optimal computational performance. The Labeled Faces in the Wild (LFW) database and self-collection face database were selected for experiments. It turns out that the true positive rate is improved by 10.97–13.24% and the acceleration ratio (the ratio between central processing unit (CPU) operation time and GPU time) is 5–6 times for the LFW face database. For the self-collection, the true positive rate increased by 12.65–15.31%, and the acceleration ratio improved by a factor of 6–7.

Fatigue is a common state of mankind characterized by a reduction in the level of consciousness and alertness. Therefore, the recognition of fatigue and sleepiness has become indispensable in many alertness-dependent situations, such as when driving vehicles on public roads, performing demanding tasks in the workplace, or monitoring intensive care unit patients. This study proposes a method based on novel multi-feature fusion to detect fatigue and sleepiness by using traditional image processing and heart rate variability (HRV). The proposed method performs initial feature extraction using InceptionV3 (a convolutional neural network (CNN)), following which the second decision is made by a long short-term memory network (LSTM) using the features collected by InceptionV3 to process the sequence of video data for recognition. The LSTM provides coherent and precise sequence recognition that avoids static distortions. Then, the final decision is made by the blood volume pulse vector (PBV) method after the features are fused. Because fatigue recognition is usually employed to monitor driver fatigue, we verified the feasibility of our method by testing its ability to successfully recognize driver fatigue. Following the experiments, we compared the different steps in the proposed method with those in existing methods. We selected four other methods to perform the comparison tests and used the same videos for training networks. In comparison with state-of-the-art methods, our method in its entirety achieved an average increase of 5% in terms of both accuracy and stability.

The convolutional neural network (CNN) has made great strides in the area of voiceprint recognition; but it needs a huge number of data samples to train a deep neural network. In practice, it is too difficult to get a large number of training samples, and it cannot achieve a better convergence state due to the limited dataset. In order to solve this question, a new method using a deep migration hybrid model is put forward, which makes it easier to realize voiceprint recognition for small samples. Firstly, it uses Transfer Learning to transfer the trained network from the big sample voiceprint dataset to our limited voiceprint dataset for the further training. Fully-connected layers of a pre-training model are replaced by restricted Boltzmann machine layers. Secondly, the approach of Data Augmentation is adopted to increase the number of voiceprint datasets. Finally, we introduce fast batch normalization algorithms to improve the speed of the network convergence and shorten the training time. Our new voiceprint recognition approach uses the TLCNN-RBM (convolutional neural network mixed restricted Boltzmann machine based on transfer learning) model, which is the deep migration hybrid model that is used to achieve an average accuracy of over 97%, which is higher than that when using either CNN or the TL-CNN network (convolutional neural network based on transfer learning). Thus, an effective method for a small sample of voiceprint recognition has been provided.

Existing vision-based fatigue detection methods commonly utilize RGB cameras to extract facial and physiological features for monitoring driver fatigue. These features often include single indicators such as eyelid movement, yawning frequency, and heart rate. However, the accuracy of RGB cameras can be affected by factors like varying lighting conditions and motion. To address these challenges, we propose a non-invasive method for multi-modal fusion fatigue detection called RPPMT-CNN-BiLSTM. This method incorporates a feature extraction enhancement module based on the improved Pan–Tompkins algorithm and 1D-MTCNN. This enhances the accuracy of heart rate signal extraction and eyelid features. Furthermore, we use one-dimensional neural networks to construct two models based on heart rate and PERCLOS values, forming a fatigue detection model. To enhance the robustness and accuracy of fatigue detection, the trained model data results are input into the BiLSTM network. This generates a time-fitting relationship between the data extracted from the CNN, allowing for effective dynamic modeling and achieving multi-modal fusion fatigue detection. Numerous experiments validate the effectiveness of the proposed method, achieving an accuracy of 98.2% on the self-made MDAD (Multi-Modal Driver Alertness Dataset). This underscores the feasibility of the algorithm. In comparison with traditional methods, our approach demonstrates higher accuracy and positively contributes to maintaining traffic safety, thereby advancing the field of smart transportation.