CCCaisen Chen

Caisen Chen

*This is an algorithmically generated author record based on publicly available information.

Caisen Chen has a total of 26 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Gear Fault Detection in a Planetary Gearbox Using Deep Belief Network

    by Hu Hao, Feng Fuzhou, Jiang Feng, Zhou Xun, Zhu Junzhen, Xue Jun, Jiang Pengcheng, Li Yazhi, Qian Yongchan, Sun Guanghui, et al.

    Traditional prognostics and health management (PHM) methods for fault detection require complex signal processing and manual fault feature extraction, and the accuracy is low. To address these problems, a fault diagnosis method of planetary gearbox based on deep belief networks (DBNs) is proposed. Firstly, the vibration signals of the planetary gearbox are collected and analyzed in the time domain and the frequency domain. Then, the DBN model and optimal parameters are determined to meet the task requirements. Finally, the vibration data is divided into training set and test set and input into the DBN model, which can realize the automatic feature extraction and fault recognition of vibration signals. The results show that the identification accuracy reaches 97% under five working conditions of planetary gearbox.


  • Open Access

    Instance Segmentation Network With Self-Distillation for Scene Text Detection

    by Peng Yang, Guowei Yang, Xun Gong, Pingping Wu, Xu Han, Jiasong Wu, Caisen Chen

    Segmentation based methods have become the mainstream for detecting scene text with arbitrary orientations and shapes. In order to address challenging problems such as separating the text instances that are very close to each other, however, these methods often require time-consuming post-processing. In this paper, we propose an instance segmentation network (ISNet), which simultaneously generates prototype masks and per-instance mask coefficients. After linearly combining the two components, ISNet can implement fast text location. Furthermore, we apply self-distillation to train the ISNet and refine its detection accuracy. We have evaluated the proposed method on four popular benchmarks, i.e., ICDAR2015, ICDAR2017 MLT, CTW1500 and Total-Text, and the experimental results show that it can achieve better tradeoff between accuracy and efficiency for scene text detection.


  • Open Access

    Temperature Compensation of Piezo-Resistive Pressure Sensor Utilizing Ensemble AMPSO-SVR Based on Improved Adaboost.RT

    by Ji Li, Chentao Zhang, Xukun Zhang, Honglin He, Wenguang Liu, Caisen Chen

    As the silicon material is severely influenced by the ambient temperature, the silicon piezo-resistive pressure sensor remarkably suffers from a strong nonlinearity in the response characteristic as the ambient temperature varies. To address this crucial issue, an adaptive mutation particle swarm optimization optimized support vector regression (AMPSO-SVR) combined with improved AdaBoost.RT algorithm is presented. The opposition-based learning initialization and Levy mutation is applied in the adaptive mutation particle swarm optimization (AMPSO) to achieve the appropriate model selection task which directly determines the performance of SVR. The performance of original AdaBoost.RT is improved by a dynamical modification approach for threshold and quoted error criterion. In order to verify the effectiveness of the proposed temperature compensation approach, several additional optimization methods such as Cuckoo search (CS), dragonfly algorithm (DA), multi-verse optimizer (MVO), conventional particle swarm optimization (PSO), Levy flight improved particle swarm optimization (Levy-PSO) and the AMPSO combined with SVR are investigated. The minimum quoted error, maximum quoted error, the mean quoted error and the variance of the quoted error over testing data obtained by the proposed method are 6.8764×10-5, 6.4463×10-4, 3.2619×10-4 and 2.5714×10-8 respectively, which are superior to the corresponding indices obtained by other methods. The analysis of simulation results indicates the method proposed in this research is applicable, effective and efficient for industrial application.


  • Power Attack and Evaluation on Elliptic Curve Cryptosystem in a Commercial Smart Card

    by Annv Liu, An Wang, Yongyin Dong, Ying Shi, Caisen Chen, Guoshuang Zhang

    Elliptic Curve Cryptography (ECC) is a widely used cryptographic algorithm in our life. Although its theory has become one of the focuses, it is difficult to ensure its security during the process of implementation. One can break it easily by using side-channel attack such as power attack. In this paper, a key recovery platform is built with computer, filter, signal acquisition module, oscilloscope, and so on. Then, we implement the simple power attack on ECC, and employ self-correlation analysis to make it easy to recover the secret key automatically. Finally, some relevant and effective countermeasures are suggested against power attacks.