Deep Learning-based Recognition Research Articles

Deep Learning-Based Recognition of Modes and Mode Groups in Multimode Optical Fiber Communication System

Deep Learning-Based Recognition of Modes and Mode Groups in Multimode Optical Fiber Communication System

Spatial Transformer Network-Based Automatic Modulation Recognition of Blind Signals

Modulation recognition of communication signals plays an important role in both civil and military uses. Neural network-based modulation recognition methods can extract high-level abstract features which can be adopted for classification of modulation types. Compared with traditional recognition methods based on manually defined features, they have the advantage of higher recognition rate. However, in actual modulation recognition scenarios, due to inaccurate estimation of receiving parameters and other reasons, the input signal samples for modulation recognition may have large phase, frequency offsets, and time scale changes. Existing deep learning-based modulation recognition methods have not considered the influences brought by the above issues, thus resulting in a decreased recognition rate. A modulation recognition method based on the spatial transformation network is proposed in this paper. In the proposed network, some prior models for synchronization in communication are introduced, and the priori models are realized through the spatial transformation subnetwork, so as to reduce the influence of phase, frequency offsets, and time scale differences. Experiments on simulated datasets prove that compared with the traditional CNN, ResNet, and the CLDNN, the recognition rate of the proposed method has increased by 8.0%, 5.8%, and 4.6%, respectively, when the signal-to-noise ratio is greater than 0. Moreover, the proposed network is also easier to train. The training time required for convergence has reduced by 4.5% and 80.7% compared to the ResNet and CLDNN, respectively.

Automated evaluation of tumor spheroid behavior in 3D culture using deep learning-based recognition

Three-dimensional in vitro tumor models provide more physiologically relevant responses to drugs than 2D models, but the lack of proper evaluation indices and the laborious quantitation of tumor behavior in 3D have limited the use of 3D tumor models in large-scale preclinical drug screening. Here we propose two indices of 3D tumor invasiveness—the excess perimeter index (EPI) and the multiscale entropy index (MSEI)—and combine these indices with a new convolutional neural network-based algorithm for tumor spheroid boundary detection. This new algorithm for 3D tumor boundary detection and invasiveness analysis is more accurate than any other existing algorithms. We apply this spheroid monitoring and AI-based recognition technique (“SMART”) to evaluating the invasiveness of tumor spheroids grown from tumor cell lines and from primary tumor cells in 3D culture.

Data-driven floor plan understanding in rural residential buildings via deep recognition

Automatic understanding of floor plan images is a key component of various applications. Due to the style diversity of rural housing design, the latest learning-based approaches cannot achieve satisfactory recognition results. In this paper, we present a new framework for parsing floor plans of rural residence that combines semantic neural networks with a post-processed room segmentation. First, we take case studies from typical residential buildings in China’s rural areas and provide a novel image dataset, called RuralHomeData, containing 800 rural residence floor plans with accurate man-made annotations. Based on the dataset, we propose a new deep learning-based recognition framework using a joint neural network to predict the geometric elements and text information on the floor plan simultaneously. Our insight is that walls and openings (doors and windows) are the basic elements corresponding to the room boundary that a closed 1D loop must form a certain room. Then the semantic information (e.g., the room function) of room regions can be obtained through text detection and identification. Furthermore, we use the MIQP algorithm to divide the area containing multiple room type texts into multiple room areas. Finally, the input floor plan can be transformed into a room layout graph with room attributes and adjacent relationships. The proposed algorithm has been tested on both urban and rural datasets, and the experimental results demonstrate our efficiency and robustness compared with the state-of-the-art methods.

Deep Learning-Based Industry 4.0 and Internet of Things towards Effective Energy Management for Smart Buildings.

Worldwide, energy consumption and saving represent the main challenges for all sectors, most importantly in industrial and domestic sectors. The internet of things (IoT) is a new technology that establishes the core of Industry 4.0. The IoT enables the sharing of signals between devices and machines via the internet. Besides, the IoT system enables the utilization of artificial intelligence (AI) techniques to manage and control the signals between different machines based on intelligence decisions. The paper’s innovation is to introduce a deep learning and IoT based approach to control the operation of air conditioners in order to reduce energy consumption. To achieve such an ambitious target, we have proposed a deep learning-based people detection system utilizing the YOLOv3 algorithm to count the number of persons in a specific area. Accordingly, the operation of the air conditioners could be optimally managed in a smart building. Furthermore, the number of persons and the status of the air conditioners are published via the internet to the dashboard of the IoT platform. The proposed system enhances decision making about energy consumption. To affirm the efficacy and effectiveness of the proposed approach, intensive test scenarios are simulated in a specific smart building considering the existence of air conditioners. The simulation results emphasize that the proposed deep learning-based recognition algorithm can accurately detect the number of persons in the specified area, thanks to its ability to model highly non-linear relationships in data. The detection status can also be successfully published on the dashboard of the IoT platform. Another vital application of the proposed promising approach is in the remote management of diverse controllable devices.

A Deep Learning-Based Recognition Technique for Plant Leaf Classification

In the practice of plant classification, the design of hand-crafted features is more dependent on the ability of computer vision experts to encode morphological characters that are predefined by botanists. However, the distinct features that each plant has as demonstrated by its leaves can be automatically learned based on the end-to-end advantage of Deep Learning algorithms. Therefore, Deep Learning based plant leaf recognition methods is an important approach nowadays. In this article, we are applying three technologies to achieve a model with high accuracy for plant classification. A Conditional Generative Adversarial Network was used to generate synthetic data, a Convolutional Neural Network was used for feature extraction and the rich extracted features were fed into a Logistic Regression classifier for efficient classification of the plant species. The effectiveness of this method can be seen in the wealth of plant datasets that it was tested on. The paper contains results on seven datasets with different modalities. We utilized both Deep Learning and Logistic regression in effectively classifying the plants using their leaf images with accuracies averaging 96.1% for about eight datasets used, but greater for the individual datasets from 99.0 to 100% on some individual datasets. Extensive experiments on each of the datasets demonstrate the superiority of our method compared with others and are highlighted in our results.

Adaptive Multiple Layer Retinex-Enabled Color Face Enhancement for Deep Learning-Based Recognition

Adaptive Multiple Layer Retinex-Enabled Color Face Enhancement for Deep Learning-Based Recognition

A Performance Evaluation of Classic Convolutional Neural Networks for 2D and 3D Palmprint and Palm Vein Recognition

Palmprint recognition and palm vein recognition are two emerging biometrics technologies. In the past two decades, many traditional methods have been proposed for palmprint recognition and palm vein recognition, and have achieved impressive results. However, the research on deep learning-based palmprint recognition and palm vein recognition is still very preliminary. In this paper, in order to investigate the problem of deep learning based 2D and 3D palmprint recognition and palm vein recognition in-depth, we conduct performance evaluation of seventeen representative and classic convolutional neural networks (CNNs) on one 3D palmprint database, five 2D palmprint databases and two palm vein databases. A lot of experiments have been carried out in the conditions of different network structures, different learning rates, and different numbers of network layers. We have also conducted experiments on both separate data mode and mixed data mode. Experimental results show that these classic CNNs can achieve promising recognition results, and the recognition performance of recently proposed CNNs is better. Particularly, among classic CNNs, one of the recently proposed classic CNNs, i.e., EfficientNet achieves the best recognition accuracy. However, the recognition performance of classic CNNs is still slightly worse than that of some traditional recognition methods.

Solar-Powered Deep Learning-Based Recognition System of Daily Used Objects and Human Faces for Assistance of the Visually Impaired

This paper introduces a novel low-cost solar-powered wearable assistive technology (AT) device, whose aim is to provide continuous, real-time object recognition to ease the finding of the objects for visually impaired (VI) people in daily life. The system consists of three major components: a miniature low-cost camera, a system on module (SoM) computing unit, and an ultrasonic sensor. The first is worn on the user’s eyeglasses and acquires real-time video of the nearby space. The second is worn as a belt and runs deep learning-based methods and spatial algorithms which process the video coming from the camera performing objects’ detection and recognition. The third assists on positioning the objects found in the surrounding space. The developed device provides audible descriptive sentences as feedback to the user involving the objects recognized and their position referenced to the user gaze. After a proper power consumption analysis, a wearable solar harvesting system, integrated with the developed AT device, has been designed and tested to extend the energy autonomy in the different operating modes and scenarios. Experimental results obtained with the developed low-cost AT device have demonstrated an accurate and reliable real-time object identification with an 86% correct recognition rate and 215 ms average time interval (in case of high-speed SoM operating mode) for the image processing. The proposed system is capable of recognizing the 91 objects offered by the Microsoft Common Objects in Context (COCO) dataset plus several custom objects and human faces. In addition, a simple and scalable methodology for using image datasets and training of Convolutional Neural Networks (CNNs) is introduced to add objects to the system and increase its repertory. It is also demonstrated that comprehensive trainings involving 100 images per targeted object achieve 89% recognition rates, while fast trainings with only 12 images achieve acceptable recognition rates of 55%.

LiDAR Point Cloud Recognition and Visualization with Deep Learning for Overhead Contact Inspection.

As overhead contact (OC) is an essential part of power supply systems in high-speed railways, it is necessary to regularly inspect and repair abnormal OC components. Relative to manual inspection, applying LiDAR (light detection and ranging) to OC inspection can improve efficiency, accuracy, and safety, but it faces challenges to efficiently and effectively segment LiDAR point cloud data and identify catenary components. Recent deep learning-based recognition methods are rarely employed to recognize OC components, because they have high computational complexity, while their accuracy needs to be improved. To track these problems, we first propose a lightweight model, RobotNet, with depthwise and pointwise convolutions and an attention module to recognize the point cloud. Second, we optimize RobotNet to accelerate its recognition speed on embedded devices using an existing compilation tool. Third, we design software to facilitate the visualization of point cloud data. Our software can not only display a large amount of point cloud data, but also visualize the details of OC components. Extensive experiments demonstrate that RobotNet recognizes OC components more accurately and efficiently than others. The inference speed of the optimized RobotNet increases by an order of magnitude. RobotNet has lower computational complexity than other studies. The visualization results also show that our recognition method is effective.

Sequential image deep learning-based Wi-Fi human activity recognition method

Sequential image deep learning-based Wi-Fi human activity recognition method

Deep learning-based recognition of cell structures in fluorescence microscopy sequences with respect to their morphology on cells infected with Marburg virus

Abstract The Institute of Virology at the Philipps-Universität Marburg is currently researching possible drugs to combat the Marburg virus. This involves classifying cell structures based on fluoroscopic microscopic image sequences. Conventionally, membranes of cells must be marked for better analysis, which is time consuming. In this work, an approach is presented to identify cell structures in images that are marked for subviral particles. It could be shown that there is a correlation between the distribution of subviral particles in an infected cell and the position of the cell’s structures. The segmentation is performed with a "Mask-R-CNN" algorithm, presented in this work. The model (a region-based convolutional neural network) is applied to enable a robust and fast recognition of cell structures. Furthermore, the network architecture is described. The proposed method is tested on data evaluated by experts. The results show a high potential and demonstrate that the method is suitable.

INDOOR NAVIGATION AND MAPPING: PERFORMANCE ANALYSIS OF UWB-BASED PLATFORM POSITIONING

Abstract. The increasing demand for reliable indoor navigation systems is leading the research community to investigate various approaches to obtain effective solutions usable with mobile devices. Among the recently proposed strategies, Ultra-Wide Band (UWB) positioning systems are worth to be mentioned because of their good performance in a wide range of operating conditions. However, such performance can be significantly degraded by large UWB range errors; mostly, due to non-line-of-sight (NLOS) measurements. This paper considers the integration of UWB with vision to support navigation and mapping applications. In particular, this work compares positioning results obtained with a simultaneous localization and mapping (SLAM) algorithm, exploiting a standard and a Time-of-Flight (ToF) camera, with those obtained with UWB, and then with the integration of UWB and vision. For the latter, a deep learning-based recognition approach was developed to detect UWB devices in camera frames. Such information is both introduced in the navigation algorithm and used to detect NLOS UWB measurements. The integration of this information allowed a 20% positioning error reduction in this case study.

Unpaired Domain Transfer for Data Augment in Face Recognition

Face recognition is one of the hottest issues in the field of computer vision and pattern recognition. Deep learning-based recognition models already have more exceptional recognition ability than the human being on open datasets, but still cannot fully undertake the identity recognition task in real scenarios without human assistance. In this paper, we mainly analyze two obstacles, i.e., domain gap and training data shortage. We propose the unpaired Domain Transfer Generative Adversarial Network (DT-GAN) to relieve these two obstacles. We improve the GAN baseline to bridge the domain gap among datasets by generating images conforming to the style of a target domain by learning the mapping between the source domain and target domain. The generator could synthesize face with an arbitrary viewpoint at the same time. The model is trained with a combination of style transfer loss, identity loss, and pose loss, which ensures the successive domain transfer and data augment. We conduct experiments to testify the effectiveness and reasonability of DT-GAN. Experimental results demonstrate the recognition performance is dramatically boosted after domain transfer and data augment.

Quantitative liver fat fraction measurement by multi-view sonography using deep learning and attention maps

Qualitative sonography is used to assess nonalcoholic fatty liver disease (NAFLD), an important health issue worldwide. We used B-mode image deep-learning to objectively assess NAFLD in 4 views of the liver (hepatic veins at confluence with inferior vena cava, right portal vein, right posterior portal vein and liver/kidney) in 135 patients with known or suspected NAFLD. Transfer learning with a deep convolutional neural network (CNN) was applied for quantifying fat fraction and diagnosing fatty liver (≥ 5%) using contemporaneous MRI-PDFF as ground truth. Single and multi-view learning approaches were compared. Class activation mapping generated attention maps to highlight regions important for deep learning-based recognition. The most accurate single view was hepatic veins, with area under the receiver operating characteristic curve (AUC) of 0.86 and Spearman’s rank correlation coefficient of 0.65. A multi-view ensemble of deep-learning models trained for each view separately improved AUC (0.93) and correlation coefficient (0.76). Attention maps highlighted regions known to be used by radiologists in their qualitative assessment, e.g., hepatic vein-parenchyma interface and liver-kidney interface. Machine learning of four liver views can automatically and objectively assess liver fat. Class activation mapping suggests that the CNN focuses on similar features as radiologists. [No. R01DK106419.]

Pulsar candidate recognition with deep learning

In this paper, we present a deep learning-based recognition algorithm to identify pulsars by observing data containing millions of candidates including radio frequency interference and noise sources. The dataset is obtained from the High Time Resolution Universe survey created and updated by the Parkes telescope. We investigate several effective single and combined features via simple logistic regression. To deal with the imbalanced dataset, we oversimplify the original dataset at different sampling rates, which is also one of the learning parameters. After training the pre-processed dataset via a convolutional neural network, we provide a cross-validated evaluation of all candidates. Results show that the deep-learning based recognition algorithm can identify the pulsar and radio frequency interference signals with high accuracy. The precision and recall of radio frequency interference are both 100%, and those of pulsars are 91% and 94%, respectively.

A deep learning-based recognition method for degradation monitoring of ball screw with multi-sensor data fusion

In this paper, a novel intelligent ball screw degradation recognition method based on deep belief networks (DBN) and multi-sensor data fusion is proposed. First, the derived method calculates frequency spectrums of raw signals, and the fused frequency spectrums are calculated by the multi-sensor data fusion. Then, a deep learning-based recognition model that can estimate the degradation condition of ball screw automatically is established with the fused dataset. The effectiveness of the proposed method is validated using dataset collected from the degradation test of ball screw. The dataset contains massive samples involving 7 degradation stages under 9 working conditions by 3 accelerometers. The classification results indicate that the proposed DBN-based method is able to mine intrinsic characteristics from the fused frequency spectrums adaptively and obtain a superior recognition accuracy. Finally, two comparative studies are performed to show the advantage of the proposed DBN-based method in ball screw degradation condition recognition.