MSIGHT Technologies (China)

In order to improve the servo accuracy and dynamic performance of permanent magnet synchronous motors (PMSM), a physics-guided neural network (PGNN) model is proposed, which can be used to establish a dynamic inverse model of PMSM-based servo systems. First, based on the characteristics of PMSM, a PGNN model is constructed, which is divided into a preprocessing model part, a physical model part, and a neural network part. Then, the above model is applied to the design of feedforward controller to improve servo performance. Considering that the dynamic characteristics of servo systems are variable, this paper further proposes an adaptive method that can periodically update model parameters based on input and output data of the system, thereby improving control effectiveness. The simulation experiment results show that the PGNN model proposed in this paper has high accuracy and good generalization ability, and can accurately describe the nonlinear characteristics of PMSM. The control effect of the feedforward controller based on this model is better than traditional modeling method and the model established in our previous work. Therefore, the control method proposed in this paper holds significant value and is crucial for advancing the intelligent development of PMSM systems.

In the fields of both vestibular and auditory research, reliable vestibular function tests are essential. However, unlike the auditory function tests, which use standard Auditory Brainstem Response (ABR) equipment, there is no equivalent widely adopted apparatus for vestibular tests. Vestibulo-ocular reflexes (VORs) are the compensatory ocular reflexes that ensure stable vision during head motion. VORs are widely used in clinics to diagnose vestibular deficits. In the research, VORs have been used by various groups to evaluate the mouse vestibular function. However, the effectiveness of VOR tests has not been systematically evaluated with appropriate mouse models, and the lack of commercial equipment hampers its accessibility, confining vestibular testing to a select few labs. In this study, we developed an integrated and surgery-free instrument system with both angular VOR (aVOR) and off-vertical axis rotation (OVAR) modes for evaluating mouse vestibular function. In addition, the eye rotation calibrations used in this study standardize the data between instruments. To demonstrate its validity and efficacy of the testing equipment, we evaluated four mouse models, including both genders, with peripheral vestibular deficits: 1) mice injected with the vestibulotoxic drug 3,3'-iminodiproprionitrile (IDPN, 2mg/g and 4mg/g, 3 male/3 female per group); 2) Critical MET-related mutant mice (Cdh23v2J/v2J, 4 male/4 female and TMC1-/-, 6 male/5 female); 3) Vestibular-specific mutant mice (Zpld1-/-, 6 male/6 female, for semicircular canal dysfunction and Otop1tlt/tlt, 3 male/2 female, for otoconia deficient); 4) Unilateral vestibular lesion (UVL) mouse model (3 male/3 female per group) where gentamicin was injected into horizontal semicircular canal. The results showed: 1) Quantification of vestibular deficits can be achieved as a daily routine; 2) Both the horizontal semicircular canal and otolith organs can be assessed, respectively; and 3) The lesion side of UVL can be identified. These test results reveal the potential of our system as standard equipment for evaluating common vestibular deficits in mice.

The high-speed railway (HSR) transportation system in China has been growing rapidly during the past decade. In 2016, the total length of HSR in China has reached to 22,000 kilometers, and there are over 2,000 pairs of high speed trains operating daily. With the advancement of design and manufacturing technologies, the reliability and construction costs have been improved significantly. However, there is still great need for reduction of their operation and maintenance costs. With such incentive, a pilot project has been launched to develop a prognostics and health management system for rolling stock to transform the maintenance paradigm from preventive to predictive maintenance. Considering the high task variety and big data environment in HSR real-time monitoring system, a cyberphysical system (CPS) architecture is proposed as the framework for its PHM system. This paper reviews the needs of predictive maintenance for the HSR system, and then present a concept design of the CPS-enabled smart operation and maintenance system.

MR images (MRIs) accurate segmentation of brain lesions is important for improving cancer diagnosis, surgical planning, and prediction of outcome. However, manual and accurate segmentation of brain lesions from 3D MRIs is highly expensive, time-consuming, and prone to user biases. We present an efficient yet conceptually simple brain segmentation network (referred as Brain SegNet), which is a 3D residual framework for automatic voxel-wise segmentation of brain lesion. Our model is able to directly predict dense voxel segmentation of brain tumor or ischemic stroke regions in 3D brain MRIs. The proposed 3D segmentation network can run at about 0.5s per MRIs - about 50 times faster than previous approaches Med Image Anal 43: 98–111, 2018, Med Image Anal 36:61–78, 2017. Our model is evaluated on the BRATS 2015 benchmark for brain tumor segmentation, where it obtains state-of-the-art results, by surpassing recently published results reported in Med Image Anal 43: 98–111, 2018, Med Image Anal 36:61–78, 2017. We further applied the proposed Brain SegNet for ischemic stroke lesion outcome prediction, with impressive results achieved on the Ischemic Stroke Lesion Segmentation (ISLES) 2017 database.

A family of loss functions built on pair-based computation have been proposed in the literature which provide a myriad of solutions for deep metric learning. In this paper, we provide a general weighting framework for understanding recent pair-based loss functions. Our contributions are three-fold: (1) we establish a General Pair Weighting (GPW) framework, which casts the sampling problem of deep metric learning into a unified view of pair weighting through gradient analysis, providing a powerful tool for understanding recent pair-based loss functions; (2) we show that with GPW, various existing pair-based methods can be compared and discussed comprehensively, with clear differences and key limitations identified; (3) we propose a new loss called multi-similarity loss (MS loss) under the GPW, which is implemented in two iterative steps (i.e., mining and weighting). This allows it to fully consider three similarities for pair weighting, providing a more principled approach for collecting and weighting informative pairs. Finally, the proposed MS loss obtains new state-of-the-art performance on four image retrieval benchmarks, where it outperforms the most recent approaches, such as ABE\cite{Kim_2018_ECCV} and HTL by a large margin: 60.6% to 65.7% on CUB200, and 80.9% to 88.0% on In-Shop Clothes Retrieval dataset at Recall@1. Code is available at https://github.com/MalongTech/research-ms-loss.

Text detection and recognition in natural images have long been considered as two separate tasks that are processed sequentially. Training of two tasks in a unified framework is non-trivial due to significant dif- ferences in optimisation difficulties. In this work, we present a conceptually simple yet efficient framework that simultaneously processes the two tasks in one shot. Our main contributions are three-fold: 1) we propose a novel text-alignment layer that allows it to precisely compute convolutional features of a text instance in ar- bitrary orientation, which is the key to boost the per- formance; 2) a character attention mechanism is introduced by using character spatial information as explicit supervision, leading to large improvements in recognition; 3) two technologies, together with a new RNN branch for word recognition, are integrated seamlessly into a single model which is end-to-end trainable. This allows the two tasks to work collaboratively by shar- ing convolutional features, which is critical to identify challenging text instances. Our model achieves impressive results in end-to-end recognition on the ICDAR2015 dataset, significantly advancing most recent results, with improvements of F-measure from (0.54, 0.51, 0.47) to (0.82, 0.77, 0.63), by using a strong, weak and generic lexicon respectively. Thanks to joint training, our method can also serve as a good detec- tor by achieving a new state-of-the-art detection performance on two datasets.

Text detection and recognition in natural images have long been considered as two separate tasks that are processed sequentially. Training of two tasks in a unified framework is non-trivial due to significant dif- ferences in optimisation difficulties. In this work, we present a conceptually simple yet efficient framework that simultaneously processes the two tasks in one shot. Our main contributions are three-fold: 1) we propose a novel text-alignment layer that allows it to precisely compute convolutional features of a text instance in ar- bitrary orientation, which is the key to boost the per- formance; 2) a character attention mechanism is introduced by using character spatial information as explicit supervision, leading to large improvements in recognition; 3) two technologies, together with a new RNN branch for word recognition, are integrated seamlessly into a single model which is end-to-end trainable. This allows the two tasks to work collaboratively by shar- ing convolutional features, which is critical to identify challenging text instances. Our model achieves impressive results in end-to-end recognition on the ICDAR2015 dataset, significantly advancing most recent results, with improvements of F-measure from (0.54, 0.51, 0.47) to (0.82, 0.77, 0.63), by using a strong, weak and generic lexicon respectively. Thanks to joint training, our method can also serve as a good detec- tor by achieving a new state-of-the-art detection performance on two datasets.

Harvesting dense pixel-level annotations to train deep neural networks for semantic segmentation is extremely expensive and unwieldy at scale. While learning from synthetic data where labels are readily available sounds promising, performance degrades significantly when testing on novel realistic data due to domain discrepancies. We present Dual Channel-wise Alignment Networks (DCAN), a simple yet effective approach to reduce domain shift at both pixel-level and feature-level. Exploring statistics in each channel of CNN feature maps, our framework performs channel-wise feature alignment, which preserves spatial structures and semantic information, in both an image generator and a segmentation network. In particular, given an image from the source domain and unlabeled samples from the target domain, the generator synthesizes new images on-the-fly to resemble samples from the target domain in appearance and the segmentation network further refines high-level features before predicting semantic maps, both of which leverage feature statistics of sampled images from the target domain. Unlike much recent and concurrent work relying on adversarial training, our framework is lightweight and easy to train. Extensive experiments on adapting models trained on synthetic segmentation benchmarks to real urban scenes demonstrate the effectiveness of the proposed framework.

This paper proposes a generic formulation that significantly expedites the training and deployment of image classification models, particularly under the scenarios of many image categories and high feature dimensions. As the core idea, our method represents both the images and learned classifiers using binary hash codes, which are simultaneously learned from the training data. Classifying an image thereby reduces to retrieving its nearest class codes in the Hamming space. Specifically, we formulate multiclass image classification as an optimization problem over binary variables. The optimization alternatingly proceeds over the binary classifiers and image hash codes. Profiting from the special property of binary codes, we show that the sub-problems can be efficiently solved through either a binary quadratic program (BQP) or a linear program. In particular, for attacking the BQP problem, we propose a novel bit-flipping procedure which enjoys high efficacy and a local optimality guarantee. Our formulation supports a large family of empirical loss functions and is, in specific, instantiated by exponential and linear losses. Comprehensive evaluations are conducted on several representative image benchmarks. The experiments consistently exhibit reduced computational and memory complexities of model training and deployment, without sacrificing classification accuracy.

The high-speed railway (HSR) transportation system in China has been growing rapidly during the past decade. In 2016, the total length of HSR in China has reached to 22,000 kilometers, and there are over 2,000 pairs of high speed trains operating daily. With the advancement of design and manufacturing technologies, the reliability and construction costs have been improved significantly. However, there is still great need for reduction of their operation and maintenance costs. With such incentive, a pilot project has been launched in CRRC to develop a prognostics and health management system for rolling stock to transform the maintenance paradigm from preventive to predictive maintenance. Considering the high task variety and big data environment in HSR real-time monitoring system, a cyber-physical system (CPS) architecture is proposed as the framework for its PHM system. This paper reviews the needs of predictive maintenance for the HSR system, and then present a concept design of the CPS-enabled smart operation and maintenance system.