Single-Stage Underwater Target Detection Based on Feature Anchor Frame Double Optimization Network.

OBJECTIVE: The shallow underwater environment is complex, with problems of color shift, uneven illumination, blurring, and distortion in the imaging process. These scenes are very unfavorable for the reasoning of the detection network. Additionally, typical object identification algorithms struggle to maintain high resilience in underwater environments due to picture domain offset, making underwater object detection problematic. METHODS: This paper proposes a single-stage detection method with the double enhancement of anchor boxes and features. The feature context relevance is improved by proposing a composite-connected backbone network. The receptive field enhancement module is introduced to enhance the multi-scale detection capability. Finally, a prediction refinement strategy is proposed, which refines the anchor frame and features through two regressions, solves the problem of feature anchor frame misalignment, and improves the detection performance of the single-stage underwater algorithm. RESULTS: We achieved an effect of 80.2 mAP on the Labeled Fish in the Wild dataset, which saves some computational resources and time while still improving accuracy. On the original basis, UWNet can achieve 2.1 AP accuracy improvement due to the powerful feature extraction function and the critical role of multi-scale functional modules. At an input resolution of 300 × 300, UWNet can provide an accuracy of 32.4 AP. When choosing the number of prediction layers, the accuracy of the four and six prediction layer structures is compared. The experiments show that on the Labeled Fish in the Wild dataset, the six prediction layers are better than the four. CONCLUSION: The single-stage underwater detection model UWNet proposed in this research has a double anchor frame and feature optimization. By adding three functional modules, the underwater detection of the single-stage detector is enhanced to address the issue that it is simple to miss detection while detecting small underwater targets.

Super-resolution reconstruction of biometric features recognition based on manifold learning and deep residual network.

BACKGROUND AND OBJECTIVE: In daily life, face information has the characteristics of uniqueness and universality. However, in a real-world scene, the image information of the face acquired by the acquisition device often contains noises such as blurring and sharpening. As such, super-resolution reconstruction of face features recognition based on manifold learning is proposed in this paper. METHODS: We reconstruct low-resolution facial expression images, introduce a simplified residual block network and manifold learning, and propose joint supervision through a new hybrid loss function, which not only retains the color and characteristics of the image, but also retains the high-frequency information. The ResNet50 network uses the weight feature of information entropy to optimize the information of the pooling layer, and the esNet50 network uses the improved PSO algorithm to optimize the initial weight of the error back-propagation phase. RESULTS: In the case of inputting extremely low resolution (6 × 6) facial expression images, the accuracy rate is increased by 9.091%. The accuracy of the high-resolution facial expressions after reconstruction with a size of 12×12 is 96.970%. The accuracy rate for happy expressions is 100%, the accuracy rate for anger, disgust, sadness, and surprise recognition is 97%, the accuracy rate for contempt is 94%, and the accuracy rate for fear is 88%. CONCLUSIONS: The experimental results verify the feasibility and superiority of the system, and effectively improve the accuracy of low-resolution facial expressions.

Data-driven trajectory tracking control for autonomous underwater vehicle based on iterative extended state observer.

In this study, we explore the precise trajectory tracking control problem of autonomous underwater vehicle (AUV) under the disturbance of the underwater environment. First, a model-free adaptive control (MFAC) is designed based on data-driven ideology and a full-form dynamic linearization (FFDL) method is utilized to online estimate time-varying parameter pseudo gradient (PG) to establish an equivalent data model of AUV motion. Second, the iterative extended state observer (IESO) scheme is designed to combine with FFDL-MFAC. Because the proposed novel controller is able to learn from repeated iterations, the proposed novel controller can estimate and compensate the model approximation error produced by external environmental unknown disturbance. Third, three-dimensional motion is decoupled into horizontal and vertical and a multi closed-loop control structure is designed that exhibits faster convergence rate and reduces sensitivity to parameter jumps than single closed-loop system. Finally, two simulation scenarios are designed featuring non external disturbance and Gaussian noise of signal-to-noise ratio of 90 dB. The simulation results reveal the superiority of FFDL. Furthermore, we adpot the technical parameters data of T-SEA I AUV to conduct numerical simulation, aunderwater trajectory as the tracking scenario and set waves to 0.5 m and current to 0.2 m/s to simulate Lv.2 ocean conditions of "International Ocean State Standard". The simulation results demonstrate the effectiveness and robustness of the proposed tracking control algorithm.

Facial expression recognition based on deep learning.

BACKGROUND AND OBJECTIVE: Facial expression recognition technology will play an increasingly important role in our daily life. Autonomous driving, virtual reality and all kinds of robots integrated into our life depend on the development of facial expression recognition technology. Many tasks in the field of computer vision are based on deep learning technology and convolutional neural network. The paper proposes an occluded expression recognition model based on the generated countermeasure network. The model is divided into two modules, namely, occluded face image restoration and face recognition. METHODS: Firstly, this paper summarizes the research status of deep facial expression recognition methods in recent ten years and the development of related facial expression database. Then, the current facial expression recognition methods based on deep learning are divided into two categories: Static facial expression recognition and dynamic facial expression recognition. The two methodswill be introduced and summarized respectively. Aiming at the advanced deep expression recognition algorithms in the field, the performance of these algorithms on common expression databases is compared, and the strengths and weaknesses of these algorithms are analyzed in detail. DISCUSSION AND RESULTS: As the task of facial expression recognition is gradually transferred from the controlled laboratory environment to the challenging real-world environment, with the rapid development of deep learning technology, deep neural network can learn discriminative features, and is gradually applied to automatic facial expression recognition task. The current deep facial expression recognition system is committed to solve the following two problems: (1) Overfitting due to lack of sufficient training data; (2) In the real world environment, other variables that have nothing to do with expression bring interference problems. CONCLUSION: From the perspective of algorithm, combining other expression models, such as facial action unit model and pleasure arousal dimension model, as well as other multimodal models, such as audio mode, 3D face depth information and human physiological information, can make expression recognition more practical.

Slicing-Tracking-Detection: Simultaneous Multi-Cylinder Detection From Large-Scale and Complex Point Clouds.

Multiple cylinders detection from large-scale and complex point clouds is a historical but challenging problem, considering the efficiency and accuracy. We propose a novel framework, named slicing-tracking-detection (STD), that detects multiple cylinders accurately and simultaneously from point clouds of large-scale and complex process plants. In this framework, the 3D cylinder detection problem is reformulated as a cylinder ingredients tracking task based on multi-object tracking (MOT). First, we generate slices from the input point cloud, and render them to slice sequence. Then, the cycle of a cylinder is modeled with a Markov Decision Process (MDP), where the ingredient is tracked with a template and the miss tracking is associated with ingredient proposals through reinforcement learning. Finally, by applying MDP for each cylinder, multiple cylinders can be detected simultaneously and accurately. Extensive experiments show that the proposed STD framework can significantly outperform the state-of-the-art approaches in efficiency, accuracy, and robustness. The source code is available at http://zhiyongsu.github.io.

Robust face recognition based on multi-task convolutional neural network.

PURPOSE: Due to the lack of prior knowledge of face images, large illumination changes, and complex backgrounds, the accuracy of face recognition is low. To address this issue, we propose a face detection and recognition algorithm based on multi-task convolutional neural network (MTCNN). METHODS: In our paper, MTCNN mainly uses three cascaded networks, and adopts the idea of candidate box plus classifier to perform fast and efficient face recognition. The model is trained on a database of 50 faces we have collected, and Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measurement (SSIM), and receiver operating characteristic (ROC) curve are used to analyse MTCNN, Region-CNN (R-CNN) and Faster R-CNN. RESULTS: The average PSNR of this technique is 1.24 dB higher than that of R-CNN and 0.94 dB higher than that of Faster R-CNN. The average SSIM value of MTCNN is 10.3% higher than R-CNN and 8.7% higher than Faster R-CNN. The Area Under Curve (AUC) of MTCNN is 97.56%, the AUC of R-CNN is 91.24%, and the AUC of Faster R-CNN is 92.01%. MTCNN has the best comprehensive performance in face recognition. For the face images with defective features, MTCNN still has the best effect. CONCLUSIONS: This algorithm can effectively improve face recognition to a certain extent. The accuracy rate and the reduction of the false detection rate of face detection can not only be better used in key places, ensure the safety of property and security of the people, improve safety, but also better reduce the waste of human resources and improve efficiency.

A Fuzzy Comprehensive CS-SVR Model-based health status evaluation of radar.

The purpose of Fuzzy Comprehensive CS-SVR Model (FCCS-SVR) is to evaluate and monitor the health status of a radar equipment and then keep its safe operation. Due to reasons such as few samples, slow changes and the nonlinear structure of data of fault monitoring signal, the health status evaluation of a radar system is quite difficult. By establishing the evaluation index system of a radar, the combination of AHP method and Entropy weight method is studied in this paper. In order to evaluate the value of health status, several optimization algorithms including PSO, GA, BA and CS are used for optimizing the parameters of SVR model. Meanwhile, in order to avoid the problem that the system is at the edge of the state, a radar health assessment method based on the combination of Fuzzy Comprehensive Evaluation and Cuckoo Search-Support Vector Regression (CS-SVR), which is named as Fuzzy Comprehensive CS-SVR (FCCS-SVR), is further proposed. The result of case analysis reflects that the state evaluation of the radar system is realized. The system performance analysis shows that the use of FCCS-SVR evaluation method provides a high recognition rate and can accurately assess the health status of the radar system.

An improved CS-LSSVM algorithm-based fault pattern recognition of ship power equipments.

A ship power equipments' fault monitoring signal usually provides few samples and the data's feature is non-linear in practical situation. This paper adopts the method of the least squares support vector machine (LSSVM) to deal with the problem of fault pattern identification in the case of small sample data. Meanwhile, in order to avoid involving a local extremum and poor convergence precision which are induced by optimizing the kernel function parameter and penalty factor of LSSVM, an improved Cuckoo Search (CS) algorithm is proposed for the purpose of parameter optimization. Based on the dynamic adaptive strategy, the newly proposed algorithm improves the recognition probability and the searching step length, which can effectively solve the problems of slow searching speed and low calculation accuracy of the CS algorithm. A benchmark example demonstrates that the CS-LSSVM algorithm can accurately and effectively identify the fault pattern types of ship power equipments.

Detecting image splicing using merged features in chroma space.

Image splicing is an image editing method to copy a part of an image and paste it onto another image, and it is commonly followed by postprocessing such as local/global blurring, compression, and resizing. To detect this kind of forgery, the image rich models, a feature set successfully used in the steganalysis is evaluated on the splicing image dataset at first, and the dominant submodel is selected as the first kind of feature. The selected feature and the DCT Markov features are used together to detect splicing forgery in the chroma channel, which is convinced effective in splicing detection. The experimental results indicate that the proposed method can detect splicing forgeries with lower error rate compared to the previous literature.