Mapping Module Research Articles

Joint communication interference system design based on parameter modulation.

In this paper, a joint communication interference integration signal waveform is proposed to satisfy the need of electronic system integration in civil and military uses, and mitigate the tension of spectrum resource. We design the system structure of the integrated signal model and propose the communication receiving processing flow of the integrated system. We utilize the dense false-target jamming style to raise the constant false alarm rate detection threshold via the delay superposition of multiple groups of frequency modulation (FM) slope mismatch jamming signals, which can play a role in protecting our target from being detected. Furthermore, linear frequency modulation (LFM) signals with different FM slopes and Doppler frequencies are obtained via the modulation mapping of communication data; thus, a single LFM signal can carry n bit data. Through correlation processing and frequency detection, code sequence information can be obtained to achieve communication function. The simulation results show that the integrated signal has the effect of shielding and jamming the pulse compression radar. Moreover, the system has a better bit error rate and a high communication rate, which can ensure that the communication task of sending accurate instructions is completed while implementing effective interference.

Safe-Nav: learning to prevent PointGoal navigation failure in unknown environments

Training robots to safely navigate (Safe-Nav) in uncertain complex environments using the RGB-D sensor is quite challenging as it involves the performance of different tasks such as obstacle avoidance, optimal path planning, and control. Traditional navigation approaches cannot generate suitable paths which guarantee enough visible features. Recent learning-based methods are still not mature enough due to their proneness to collisions and prohibitive computational cost. This paper focuses on generating safe trajectories to the desired goal while avoiding collisions and tracking failure in unknown complex environments. We present Safe-Nav, a hierarchical framework composed of the visual simultaneous localization and mapping (SLAM) module, the global planner module and the local planner module. The visual SLAM module generates the navigation map and the robot pose. The global planner module plans a local waypoint on the real-time navigation map. In the local planner module, a deep-reinforcement-learning-based (DRL-based) policy is presented for taking safe actions towards local waypoints. Our DRL-based policy can learn different navigation skills (e.g., avoiding collisions and avoiding tracking failure) through specialized modes without any supervisory signals when the PointGoal-navigation-specied reward is provided. We have demonstrated the performance of our proposed Safe-Nav in the Habitat simulation environment. Our approach outperforms the recent learning-based method and conventional navigation approach with relative improvements of over 205% (0.55 vs. 0.18) and 139% (0.55 vs. 0.23) in the success rate, respectively.

A versatile inversion approach for space/temperature/time-related thermal conductivity via deep learning

Identifying the thermophysical properties of unknown material through the measurement of temperature is of great significance in computational heat transfer. Existing numerical algorithms are generally computationally cumbersome and resource demanding. Rapid advances in deep learning (DL) offer an alternative pathway to speed up the inversion process by fully utilizing the parallel computing ability of Graphics Processing Units (GPUs). In this paper, a DL framework is proposed to reconstruct the thermal conductivity related to space, temperature or time. The whole framework consists of a forward data generation module, a denoising module and an inversion module. It is noteworthy that the physics informed neural network (PINN) is employed in the process of generating training data, which avoids the use of commercial software based on traditional methods. In order to simulate the measurement error in practical scenarios, a certain intensity of Gaussian noise is added to the generated data. After denoising by the U-net, the measured temperature is fed to the nonlinear mapping module (NMM) for the inversion of the unknown thermal conductivity. As a result, a well-trained framework can realize high precision real-time inversion even with intensive environmental noise, offering great potential for applications pertaining to the reconstruction of thermophysical properties.

Short Analog Fountain Code With Quasi-Gray Constellation Mapping Modulation Towards uRLLC

In this paper, a capacity-approaching rateless coded modulation towards ultra-reliable low latency communications (uRLLC) is proposed, including the joint design of short block-length analog fountain code (AFC) and quasi-gray constellation mapping (QGCM) modulation at the transmitter, and a limited feedback mechanism for the receiver. First, based on the nonlinear constellation and quasi-gray mapping, we design a novel rateless coded modulation scheme called AFC-QGCM, and prove it can approach to the Shannon capacity. Further, the theoretical evaluation framework of information density between the achievable rate and signal-to-noise (SNR) of the AFC-QGCM schemes is derived under short block-length. Based on the analysis, we propose a rateless transmission scheme with a limited feedback link over the fading channel without precise channel state information (CSI) at the transmitter, which could utilize the theoretical analysis to determine the optimal coding and modulation parameters during the transmission. Both analysis and simulation results shown that the proposed scheme could approaching to the Polyanskiy-Poor-Verdu (PPV) meta converse benchmark in a wide range SNRs.

Millimeter Wave Wireless Assisted Robot Navigation With Link State Classification

The millimeter wave (mmWave) bands have attracted considerable attention for high precision localization applications due to the ability to capture high angular and temporal resolution measurements. This paper explores mmWave-based positioning for a target localization problem where a fixed target broadcasts mmWave signals and a mobile robotic agent attempts to capture the signals to locate and navigate to the target. A three-stage procedure is proposed: First, the mobile agent uses tensor decomposition methods to detect the multipath channel components and estimate their parameters. Second, a machine-learning trained classifier is then used to predict the <i>link state</i>, meaning if the strongest path is line-of-sight (LOS) or non-LOS (NLOS). For the NLOS case, the link state predictor also determines if the strongest path arrived via one or more reflections. Third, based on the link state, the agent either follows the estimated angles or uses computer vision or other sensor to explore and map the environment. The method is demonstrated on a large dataset of indoor environments supplemented with ray tracing to simulate the wireless propagation. The path estimation and link state classification are also integrated into a state-of-the-art neural simultaneous localization and mapping (SLAM) module to augment camera and LIDAR-based navigation. It is shown that the link state classifier can successfully generalize to completely new environments outside the training set. In addition, the neural-SLAM module with the wireless path estimation and link state classifier provides rapid navigation to the target, close to a baseline that knows the target location.

A Cascaded Spectral–Spatial CNN Model for Super-Resolution River Mapping With MODIS Imagery

Rivers are important elements of the earth’s ecosystem, and their spatial distribution information is critical for the study of hydrological and biogeochemical processes. Moderate Resolution Imaging Spectroradiometer (MODIS) imagery has been widely used for river mapping due to its high temporal resolution and long-term observation records, which are essential to capture the rapid fluctuation of rivers. However, when the conventional hard classification methods are used, the accuracy of the river maps produced from MODIS data (and especially for those rivers with narrow widths) is often limited because mixed pixels are common in MODIS imagery, due to coarse spatial resolution. In this article, a cascaded spectral–spatial information combined deep convolutional neural network (CNN) model for super-resolution river mapping (DeepRivSRM) is proposed to produce Landsat-like fine-resolution river maps from MODIS images. In DeepRivSRM, a CNN made up of a spectral unmixing module and a super-resolution mapping (SRM) module is introduced to handle the spectral and spatial information simultaneously. Moreover, for training the DeepRivSRM model, an adaptive cross-entropy loss function incorporating the fraction information of the rivers is designed to improve the performance of the DeepRivSRM model for small rivers. The proposed method was evaluated with MODIS images from three test sites and was compared with hard classification, a conventional SRM method, and a CNN-based SRM method. The results show that DeepRivSRM can generate more accurate river maps by effectively learning the subpixel-scale spatial–spectral information in MODIS imagery.

CVIDS: A Collaborative Localization and Dense Mapping Framework for Multi-Agent Based Visual-Inertial SLAM

Nowadays, visual SLAM (Simultaneous Localization And Mapping) has become a hot research topic due to its low costs and wide application scopes. Traditional visual SLAM frameworks are usually designed for single-agent systems, completing both the localization and the mapping with sensors equipped on a single robot or a mobile device. However, the mobility and work capacity of the single agent are usually limited. In reality, robots or mobile devices sometimes may be deployed in the form of clusters, such as drone formations, wearable motion capture systems, and so on. As far as we know, existing SLAM systems designed for multi-agents are still sporadic, and most of them have non-negligible limitations in functions. Specifically, on one hand, most of the existing multi-agent SLAM systems can only extract some key features and build sparse maps. On the other hand, schemes that can reconstruct the environment densely cannot get rid of the dependence on depth sensors, such as RGBD cameras or LiDARs. Systems that can yield high-density maps just with monocular camera suites are temporarily lacking. As an attempt to fill in the research gap to some extent, we design a novel collaborative SLAM system, namely CVIDS (Collaborative Visual-Inertial Dense SLAM), which follows a centralized and loosely coupled framework and can be integrated with any existing Visual-Inertial Odometry (VIO) to accomplish the co-localization and the dense reconstruction. Integrating our proposed robust loop closure detection module and two-stage pose-graph optimization pipeline, the co-localization module of CVIDS can estimate the poses of different agents in a unified coordinate system efficiently from the packed images and local poses sent by the client-ends of different agents. Besides, our motion-based dense mapping module can effectively recover the 3D structures of selected keyframes and then fuse their depth information to the global map for reconstruction. The superior performance of CVIDS is corroborated by both quantitative and qualitative experimental results. To make our results reproducible, the source code has been released at https://cslinzhang.github.io/CVIDS.

Homogeneous Aggregation Convolution for Building Extraction From Remote Sensing Images

An increasing number of convolutional neural networks are being applied to various fields, and they have achieved excellent performance. However, standard convolution cannot recognize the connection between surrounding features and cannot characterize them efficiently. For example, buildings in urban remote sensing images exhibit geometric changes, such as rotation, scaling, and local changes. The concept of dynamic convolution is proposed to solve the aforementioned problem. Existing dynamic convolution methods enhance an expression by dynamically changing the sampling points or weights of convolution. However, these end-to-end training methods do not consider which sampling points are important. In this letter, we propose a homogeneous aggregation convolution (HAC) that gives more attention to the sampling points that belong to the same class as the target point. A generate probability map module is designed to generate a probability map between target and sampling points and share this probability map across convolution layers to save computational cost. Experimental results demonstrate that the proposed HAC outperforms standard convolution, and the intersection over union and F1 score are higher than the standard convolution by 2.23% and 1.28%, respectively, on the WHU and Austin datasets. Compared with other convolutions, the proposed HAC convolution is the most efficient in building extraction.

IDLN: Iterative Distribution Learning Network for Few-Shot Remote Sensing Image Scene Classification

Few-shot remote sensing image scene classification has gained more attention due to its ability to recognize novel categories with several annotated samples. However, it is a great challenge to extract category characteristics with insufficient labeled samples. To address this issue, an iterative distribution learning network (IDLN) is proposed for the scene classification of few-shot remote sensing images. Specifically, the proposed model is a cyclic iterative architecture, which is composed of three modules to enhance the classification performance. In each iteration, similarity distribution learning module is proposed to calculate feature relations among instances first, then label mapping module is developed as the few-shot classifier based on instructive knowledge, and, finally, the attention-based feature calibration module is proposed to modify features based on label relations, and the calibrated features are imported to the next iteration. Experimental results on two public remote sensing datasets demonstrate that the proposed network is able to achieve superior performance for few-shot remote sensing image scene classification.

Learning-Aided Beam Prediction in mmWave MU-MIMO Systems for High-Speed Railway

The problem of beam alignment and tracking in high mobility scenarios such as high-speed railway (HSR) becomes extremely challenging, since large overhead cost and significant time delay are introduced for fast time-varying channel estimation. To tackle this challenge, we propose a learning-aided beam prediction scheme for HSR networks, which predicts the beam directions and the channel amplitudes within a period of future time with fine time granularity, using a group of observations. Concretely, we transform the problem of high-dimensional beam prediction into a two-stage task, i.e., a low-dimensional parameter estimation and a cascaded hybrid beamforming operation. In the first stage, the location and speed of a certain terminal are estimated by maximum likelihood criterion, and a data-driven data fusion module is designed to improve the final estimation accuracy and robustness. Then, the probable future beam directions and channel amplitudes are predicted, based on the HSR scenario priors including deterministic trajectory, motion model, and channel model. Furthermore, we incorporate a learnable non-linear mapping module into the overall beam prediction to allow non-linear tracks. Both of the proposed learnable modules are model-based and have a good interpretability. Compared to the existing beam management scheme, the proposed beam prediction has (near) zero overhead cost and time delay. Simulation results verify the effectiveness of the proposed scheme.

Graph SLAM-Based 2.5D LIDAR Mapping Module for Autonomous Vehicles

This paper proposes a unique Graph SLAM framework to generate precise 2.5D LIDAR maps in an XYZ plane. A node strategy was invented to divide the road into a set of nodes. The LIDAR point clouds are smoothly accumulated in intensity and elevation images in each node. The optimization process is decomposed into applying Graph SLAM on nodes’ intensity images for eliminating the ghosting effects of the road surface in the XY plane. This step ensures true loop-closure events between nodes and precise common area estimations in the real world. Accordingly, another Graph SLAM framework was designed to bring the nodes’ elevation images into the same Z-level by making the altitudinal errors in the common areas as small as possible. A robust cost function is detailed to properly constitute the relationships between nodes and generate the map in the Absolute Coordinate System. The framework is tested against an accurate GNSS/INS-RTK system in a very challenging environment of high buildings, dense trees and longitudinal railway bridges. The experimental results verified the robustness, reliability and efficiency of the proposed framework to generate accurate 2.5D maps with eliminating the relative and global position errors in XY and Z planes. Therefore, the generated maps significantly contribute to increasing the safety of autonomous driving regardless of the road structures and environmental factors.

FPCR-Net: Feature pyramidal correlation and residual reconstruction for optical flow estimation

Optical flow estimation is a challenging problem in the field of video analytics yet. Features of different semantics levels in a convolutional neural network (CNN) provide information of different granularity. To exploit such flexible and comprehensive information, we propose a Feature Pyramidal Correlation and Residual Reconstruction Network (FPCR-Net) for optical flow estimation from frame pairs. It consists of two main modules: pyramid correlation mapping and residual reconstruction. The pyramid correlation mapping module takes advantage of the multi-scale correlations of global/local representation by aggregating features of different scales to form a multi-level cost volume. The residual reconstruction module aims at reconstructing the sub-band high-frequency residuals of finer optical flow at each stage. Based on the pyramid correlation mapping, we further propose a correlation-warping-normalization (CWN) module to efficiently exploit the correlation dependency. Furthermore, considering the characteristics of flow warping and alignment, we integrate unsupervised and supervised losses to explore the implicit relevance and explicit constraint. Experimental results show that the proposed network achieves the promising performance for two-frame-based optical flow estimation on the challenging Sintel and KITTI 2012/2015 datasets.

Noncontact thermal mapping method based on local temperature data using deep neural network regression

Temperature monitoring of electronic devices is important to prevent the active components from overheating. In this study, a novel method to obtain the overall temperature map of electronic modules with limited local temperature data was suggested using a deep-neural-network-based multi-output regression model. To predict the entire temperature distribution with minimum data, one to six random inputs considering their diverse arrangements were applied and compared by calculating the mean absolute error. In addition, the temperature prediction accuracy of the heating element was considered as an important parameter for the performance score. Consequently, a temperature prediction accuracy of ∼96.7% was realized using three input local data points close to the heat source. Furthermore, with the other three temperature data points away from the heat source, the score increased by ∼11.6% (∼79.9 to ∼89.2%) after the hyperparameter tuning processes. These results support the precise noncontact virtual sensing technology of temperature monitoring methods for various industries, such as electric vehicles, cold-chain warehouses, and robotics.

Counting method for cultured fishes based on multi-modules and attention mechanism

Fish counting is of great significance for aquaculture companies to formulate feeding strategies and management plans. Preceding computer vision methods are difficult to solve the counting problem when the fish body is seriously overlapped and the shape changes greatly. In this research, a deep learning network model based on multi-modules and attention mechanism (MAN) is proposed to realize the counting of cultured fish which consists of feature extraction module, attention module, and density map estimation module. Among them, the feature extraction module is composed of three parallel convolutional networks which is used to extract the general feature map of the image and serves as the input of the subsequent module. The trunk convolutional network and the attention network are connected in parallel to form the attention module which is more precisely identify critical information of image in dense counting process. Thereinto, the attention network integrated a residual block without batch normalization and a convolutional layer. The distribution and the number of fishes in the image is represented by density map estimation module. In order to improve the counting accuracy and retain more details of the image, the sum of squared error (SSE) and structural similarity index (SSIM) are applied to form the loss function and train the model. In order to explore the generalization ability of the model, data from two tanks were used to validate the model. The experimental results for MAN showed that the counting accuracy is about 97.12%, and the deviation is 3.67. Compared with the MCNN and CNNs, the accuracy of MAN has been improved by about 1.95% and 2.76%. At the same time, the deviation of MAN has also been reduced by 30.23% and 41.09%, respectively. For the images of another tank, the counting accuracy of MAN is 94.69%, while the MCNN and CNNs cannot meet the application requirements of fish counting. Furthermore, the accuracy for MAN between the different distribution intervals of the number of fishes is fluctuating less. On the whole, the model not only has high counting accuracy, but also has strong stability and generalization performance. The method that based on multi-modules and attention mechanism can promote the exploration of the technology related to dense fish counting, and provide a feasible solution for the application of this technology in practical engineering.

Panoramic Visual-Inertial SLAM Tightly Coupled with a Wheel Encoder

This paper presents a panoramic visual-inertial simultaneous localization and mapping (SLAM) system that is tightly coupled with a wheel encoder, which can be used in a mobile mapping system (MMS), robot, or driverless car. The whole SLAM system is made up of four modules: 1) measurement preprocessing; 2) system initialization; 3) tracking; and 4) local mapping. The core of the system is the combined adjustment we propose for the observations of the panoramic camera, inertial measurement unit (IMU), and wheel encoder, to optimize the system state, which includes the rotation, translation, velocity, IMU bias, and local map point coordinates in the initialization, tracking, and local mapping modules. In the preprocessing, the most important contribution is that we derive the wheel preintegration based on a two-wheel differential model, which combines the many wheel measurements between two frames into a single relative motion constraint. A novel initialization algorithm is also introduced, which can be divided into two steps: 1) the first step is to initialize the visual scale and parameters of the IMU; and 2) the second step is to perform the combined adjustment for the camera, IMU, and preintegrated wheel encoder data, to refine the initial parameters. The proposed panoramic camera-IMU-wheel SLAM (PIW-SLAM) system can achieve high-precision and robust localization in challenging scenes through multi-sensor fusion. We validated the performance of the proposed system on seven different challenging data sequences from the University of Michigan North Campus Long-Term Vision and LiDAR Dataset (NCLT) dataset. The results showed that the PIW-SLAM system has a higher localization accuracy than the other state-of-the-art SLAM systems, and it also showed a superior robustness in various complex environments. We also verified the accuracy of the scale obtained by the initialization algorithm. Furthermore, we confirmed the performance of the proposed PIW-SLAM system in a wide-baseline scene of the ground motion measurement system.

Automatic segmentation of organs-at-risks of nasopharynx cancer and lung cancer by cross-layer attention fusion network with TELD-Loss.

Radiotherapy is one of the main treatments of nasopharyngeal cancer (NPC) and lung cancer. Accurate segmentation of organs at risks (OARs) in CT images is a key step in radiotherapy planning for NPC and lung cancer. However, the segmentation of OARs is influenced by the highly imbalanced size of organs, which often results in very poor segmentation results for small and difficult-to-segment organs. In addition, the complex morphological changes and fuzzy boundaries of OARs also pose great challenges to the segmentation task. In this paper, we propose a cross-layer attention fusion network (CLAF-CNN) to solve the problem of accurately segmentingOARs. In CLAF-CNN, we integrate the spatial attention maps of the adjacent spatial attention modules to make the segmentation targets more accurately focused, so that the network can capture more target-related features. In this way, the spatial attention modules in the network can be learned and optimized together. In addition, we introduce a new Top-K exponential logarithmic Dice loss (TELD-Loss) to solve the imbalance problem in OAR segmentation. The TELD-Loss further introduces a Top-K optimization mechanism based on Dice loss and exponential logarithmic loss, which makes the network pay more attention to small organs and difficult-to-segment organs, so as to enhance the overall performance of the segmentationmodel. We validated our framework on the OAR segmentation datasets of the head and neck and lung CT images in the StructSeg 2019 challenge. Experiments show that the CLAF-CNN outperforms the state-of-the-art attention-based segmentation methods in the OAR segmentation task with average Dice coefficient of 79.65% for head and neck OARs and 88.39% for lungOARs. This work provides a new network named CLAF-CNN which contains cross-layer spatial attention map fusion architecture and TELD-Loss for OAR segmentation. Results demonstrated that the proposed method could obtain accurate segmentation results for OARs, which has a potential of improving the efficiency of radiotherapy planning for nasopharynx cancer and lung cancer.

Secure OFDM-PON Using Chaotic Constellation Mapping and Probabilistic Shaping

A physical-layer secure data encryption algorithm is proposed and demonstrated using chaotic constellation mapping associated with probabilistic shaping (PS) for orthogonal frequency division multiplexing passive optical networks (OFDM-PONs). The constellation of the encrypted data is uniformly distributed after chaotic diffusion in quadratic-amplitude modulation (QAM) mapping, to eliminate the statistical characteristics, where a key space of ~10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">231</sup> is created by a 4-dimensional hyper digital chaos. Moreover, PS is introduced to improve the OFDM signal transmission. A 9.4 Gb/s encrypted PS-16-QAM OFDM signal transmission is successfully demonstrated over 20 km standard single-mode fiber (SSMF), where the gain of receiver sensitivity is ~2 dB (bit-error rate @ 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> ) using PS.

Gaussian distribution of average photon energy and spectral gain and loss of several-type photovoltaic modules at different outdoor sites around the world

Distributions of actual average photon energy (APE) at several outdoor locations around the world are investigated and can be characterized using Gaussian distribution with average APE (μ) and standard deviation (σ) values. For the comparison among different outdoor sites, their APE distributions must be calculated based on the measured solar spectra in the same wavelength range. The method to convert APE values based on different wavelength ranges into those based on same wavelength range is consequently proposed. It is shown that spectral gain and loss (G&L) of different-type photovoltaic (PV) modules calculated from actual APE distribution and its Gaussian fitted distribution are almost similar. General G&L contour maps of the different-type PV modules as functions of μ and σ values are then calculated. These contour maps readily allow us to quantitatively evaluate various-type test PV modules at several outdoor locations around the world, when the μ and σ values are known.

A network of core and subtype-specific gene expression programs in myositis.

Myositis comprises a heterogeneous group of skeletal muscle disorders which converge on chronic muscle inflammation and weakness. Our understanding of myositis pathogenesis is limited, and many myositis patients lack effective therapies. Using muscle biopsy transcriptome profiles from 119 myositis patients (spanning major clinical and serological disease subtypes) and 20 normal controls, we generated a co-expression network of 8101 dynamically regulated transcripts. This network organized the myositis transcriptome into a map of gene expression modules representing interrelated biological processes and disease signatures. Universally myositis-upregulated network modules included muscle regeneration, specific cytokine signatures, the acute phase response, and neutrophil degranulation. Universally myositis-suppressed pathways included a specific subset of myofilaments, the mitochondrial envelope, and nuclear isoforms of the anti-apoptotic humanin protein. Myositis subtype-specific modules included type 1 interferon signaling and titin (dermatomyositis), RNA processing (antisynthetase syndrome), and vasculogenesis (inclusion body myositis). Importantly, therapies exist to target influential proteins in many myositis-dysregulated modules, and nearly all modules contained understudied proteins and non-coding RNAs - many of which were extraordinarily dysregulated in myositis and may represent novel therapeutic targets. Finally, we apply our network to patient classification, finding that a deep learning algorithm trained on patient-level network "images" successfully assigned patients to clinical groups and further into molecular subclusters. Altogether, we provide a global resource to probe and contextualize differential gene expression in myositis.

A real-time ultra-low light color imaging system based on FPGA

This article shows a low light color image acquisition system, The core components of the system are the Fairchild’s SCMOS image sensor CIS1910F1111 and XILINX’s Artix-7 XC7A100T-2CSG324I FPGA, the remarkable advantage of the system is that it can obtain better color imaging effect under lower illumination environment, and the image noise is much less than other similar products. Based on the excellent imaging performance of the image detector, a high performance real-time low-light level color imaging system is developed. This imaging system can obtain the characteristic information of the targets under ultra-low illuminance environment, including the details, colors and so on. The hardware of the low light level imaging system mainly contains a color SCMOS image sensor and a FPGA, a driving circuit of a combination of DDR3, the ultra-low noise power conversion circuit and a Camera-Link and a 3G-SDI interface circuits. The SCMOS chip is used for photoelectric conversion of the shot scene and the FPGA is used for the control of the whole imaging system, image acquisition and image processing, etc, The FPGA software system consists of SCMOS initialize configuration and timing control module, automatic exposure control module, real-time color image processing module, imaging tone mapping module, image denoising module and image enhancement module. The automatic exposure control (AEC) module adaptively adjusts the average gray value of the region of interest. The module automatically calculates the exposure time and gain value of the next frame according to the current frame image data value. The real-time color image processing module includes color restoration, automatic white balance and color spaces conversion, etc. The image denoising module uses the advanced real-time guide-filter algorithm. The image tone mapping module and enhancement module are proposed based on an improved automatic threshold logarithmic and enhancement algorithm. Combining the hardware and FPGA soft algorithm with excellent performance, the imaging results show that the system can get good color image effect of the ultra-low light level about 10-2lx.