- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Ship Discrimination with Deep Convolutional Neural Networks in Sar Images
摘要： With the advantages of all-time, all-weather, and wide coverage, synthetic aperture radar (SAR) systems are widely used for ship detection to ensure marine surveillance. However, the azimuth ambiguity and buildings exhibit similar scattering mechanisms of ships, which cause false alarms in the detection of ships. To address this problem, self-designed deep convolutional neural networks with the capability to automatically learn discriminative features is applied in this paper. Two datasets, including one dataset reconstructed from IEEEDataPort SARSHIPDATA and the other constructed from 10 scenes of Sentinel-1 SAR images, are used to evaluate our approach. Experimental results reveal that our model achieves more than 95% classification accuracy on both datasets, demonstrating the effectiveness of our approach.
关键词： ship discrimination,Sentinel-1 images,synthetic aperture radar,deep convolutional neural networks
更新于2025-09-23 15:23:52
-
DeeptransMap: a considerably deep transmission estimation network for single image dehazing
摘要： Due to the ill-posed phenomenon of the classical physical model, single image dehazing based on the model has been a challenging vision task. In recent years, applying machine learning techniques to estimate a critical parameter transmission has proven to be an effective solution to this issue. Accordingly, the robustness and accuracy of learning-based transmission estimation model is extremely important, since it does impact on the final dehazing effects. The state-of-the-art dehazing algorithms by this means generally use haze-relevant features as the single input to their transmission estimation models. However, the used haze-relevant features sometimes are not sufficient and reliable in holding real intrinsic information related to haze due to their two shortcomings and ultimately bring about their less effectiveness for some dehazing cases. Based on related efforts on representation learning and deep convolutional neural networks, in this paper, we seek to achieve the robustness and accuracy of transmission estimation model for bolstering the effectiveness of single image dehazing. Specifically, we propose a hybrid model combining unsupervised and supervised learning in a considerably deep neural networks architecture, in order to achieve accurate transmission map from a single image. Experimental results demonstrate that our work performs favorably against several state-of-the-art dehazing methods with the same estimated goal and keeps efficient in terms of the computational complexity of transmission estimation.
关键词： Feature learning,Deep convolutional neural networks (CNNs),Image dehazing,Transmission estimation
更新于2025-09-23 15:23:52
-
FusionCNN: a remote sensing image fusion algorithm based on deep convolutional neural networks
摘要： In remote sensing image fusion field, traditional algorithms based on the human-made fusion rules are severely sensitive to the source images. In this paper, we proposed an image fusion algorithm using convolutional neural networks (FusionCNN). The fusion model implicitly represents a fusion rule whose inputs are a pair of source images and the output is a fused image with end-to-end property. As no datasets can be used to train FusionCNN in remote sensing field, we constructed a new dataset from a natural image set to approximate MS and Pan images. In order to obtain higher fusion quality, low frequency information of MS is used to enhance the Pan image in the pre-processing step. The method proposed in this paper overcomes the shortcomings of the traditional fusion methods in which the fusion rules are artificially formulated, because it learns an adaptive strong robust fusion function through a large amount of training data. In this paper, Landsat and Quickbird satellite data are used to verify the effectiveness of the proposed method. Experimental results show that the proposed fusion algorithm is superior to the comparative algorithms in terms of both subjective and objective evaluation.
关键词： Convolutional neural networks,Deep learning,Remote sensing image fusion,Image enhancement
更新于2025-09-23 15:23:52
-
Cuff-less continuous measurement of blood pressure using wrist and fingertip photo-plethysmograms: Evaluation and feature analysis
摘要： Continuous monitoring of blood pressure improves prevention and control of cardiovascular diseases. Currently, cuff-based oscillometric sphygmomanometers are commonly used to monitor the systolic and diastolic blood pressure. However, this technique is discontinuous in nature and inconvenient for repeated measurements. Here we have proposed indirect measurement of blood pressure from photo-plethysmograms (PPG) simultaneously recorded from wrist and fingertip. The signals were recorded from 111 participants and different morphological features were obtained from PPG and its second derivative, acceleration plethysmograms (APG). Moreover, different measures of pulse transit time (PTT) and pulse wave velocity (PWV) were obtained from the recorded PPGs. Multi-layer Neural Networks were used to estimate the non-linear relationship between these features and systolic and diastolic blood pressures (SBP and DBP). Mean absolute errors of 6.77 and 4.82 mmHg were achieved in comparison with measurements from a validated commercial oscillometric sphygmomanometer. Feature analysis provided insight about the importance of features for estimating BP, and demonstrated that these features are not the same for SBP and DBP. Using the highest-ranked 15 and 13 features obtained from moving-backward algorithm the mean absolute errors were reduced to 5.31 and 4.62 mmHg for SBP and DBP. However, the optimum optimal feature sets provided by a genetic algorithm for estimating SBP/DBP led to the lowest mean absolute errors of 4.94/4.03. These results compared to previous studies and the available standards suggest that the method is a promising substitute for oscillometric sphygmomanometers which can be used conveniently for continuous monitoring of blood pressure.
关键词： Genetic algorithms,Non-obstructive blood pressure measurement,Multi-layer neural networks,Photo-plethysmography
更新于2025-09-23 15:23:52
-
[IEEE 2018 25th IEEE International Conference on Image Processing (ICIP) - Athens, Greece (2018.10.7-2018.10.10)] 2018 25th IEEE International Conference on Image Processing (ICIP) - High-Quality Virtual View Synthesis for Light Field Cameras Using Multi-Loss Convolutional Neural Networks
摘要： Although light field cameras record both spatial and angular information, their angular and spatial resolutions are limited when capturing light field data. Thus, it is required to synthesize virtual views. In this paper, we propose high-quality virtual view synthesis based on multi-loss convolutional neural networks (CNN). We adopt multi-loss function for view synthesis in both pixel and feature spaces to increase the angular resolution of light field data. We combine three losses of feature loss, edge loss, and mean squared error (MSE) loss into the multi-loss function. We learn the view synthesis function based on simple three layers of CNN. Experimental results show that the proposed method successfully produces virtual views from light field data and outperforms state-of-the-arts in terms of peak-signal-to-noise ratio (PSNR) and structural similarity (SSIM).
关键词： Convolutional neural networks,loss function,virtual view synthesis,multi-loss,light field
更新于2025-09-23 15:23:52
-
Ship detection in spaceborne infrared images based on Convolutional Neural Networks and synthetic targets
摘要： Automatic detection of ships in spaceborne infrared images is important for both military and civil applications due to its all-weather detection capability. While deep learning methods have made great success in many image processing fields recently, training deep learning models still relies on large amount of labeled data, which may limit its application performance for infrared images target detection tasks. Considering that, we propose a new infrared ship detection method based on Convolutional Neural Networks (CNN) which is trained only with synthetic targets. For the problem of limited infrared training data, we design a Transfer Network (T-Net) to generate large amount of synthetic infrared-style ship targets from Google Earth images. The experiments are conducted on a near infrared band image (0:845μm s 0:885μm), a short wavelength infrared band image (1:560μm s 1:66μm) and a long wavelength infrared band image (2:1μm s 2:3μm) of Landsat-8 satellite. The results demonstrate the effectiveness of the target generation ability of T-Net. With only synthetic training samples, our detection method achieves a higher accuracy than other classical ship detection methods.
关键词： Convolutional Neural Networks,Spaceborne infrared image,Synthetic targets,Ship detection
更新于2025-09-23 15:23:52
-
A top-down approach for semantic segmentation of big remote sensing images
摘要： The increasing amount of remote sensing data has opened the door to new challenging research topics. Nowadays, significant efforts are devoted to pixel and object based classification in case of massive data. This paper addresses the problem of semantic segmentation of big remote sensing images. To do this, we proposed a top-down approach based on two main steps. The first step aims to compute features at the object-level. These features constitute the input of a multi-layer feed-forward network to generate a structure for classifying remote sensing objects. The goal of the second step is to use this structure to label every pixel in new images. Several experiments are conducted based on real datasets and results show good classification accuracy of the proposed approach. In addition, the comparison with existing classification techniques proves the effectiveness of the proposed approach especially for big remote sensing data.
关键词： Neural networks,Remote sensing images,Big data,Semantic segmentation
更新于2025-09-23 15:23:52
-
Artificial neural network modeling of a pilot plant jet-mixing UV/hydrogen peroxide wastewater treatment system
摘要： This study deals with the modeling and simulation of an efficient pilot plant photo-chemical wastewater treatment reactor. Treatment of an azo dye (i.e. direct red 23) was performed using a UV/H2O2 process in a jet mixing photo-reactor with 10-L volume. To model the reactor and simulate the treatment process, six important, influential physical and chemical factors such as nozzle angle (hN), nozzle diameter (dN), flow-rate (Q), irradiation time (t), H2O2 initial concentration ([H2O2]0), and pH, were taken into account. In this regard, artificial neural networks (ANNs) were employed as a powerful modeling methodology. Six different ANN architectures were constructed and most appropriate numbers for hidden neuron and learning iteration were determined based on minimization of the mean square error (MSE) function related to the testing data sets. Furthermore, simulation of the reactor efficiency, as well as sensitivity analysis, was performed via the cross-validation outputs. It was found that a three-layered feed-forward ANN composes ten hidden neurons, calibrated at 100th iteration using “trainlm” as learning algorithm and “tansig” and “purelin” as transfer functions in the hidden and output layers can model the process as the best case. The order of importance for variation of the key factors were indicated as [H2O2]0 > t > pH > Q > hN > dN.
关键词： dyes,simulation,wastewater treatment,Advance oxidation process,neural networks,photodegradation,batch reactor
更新于2025-09-23 15:23:52
-
[IEEE 2018 25th IEEE International Conference on Image Processing (ICIP) - Athens, Greece (2018.10.7-2018.10.10)] 2018 25th IEEE International Conference on Image Processing (ICIP) - PAC-Net: Pairwise Aesthetic Comparison Network for Image Aesthetic Assessment
摘要： Image aesthetic assessment is important for finding well taken and appealing photographs but is challenging due to the ambiguity and subjectivity of aesthetic criteria. We develop the pairwise aesthetic comparison network (PAC-Net), which consists of two parts: aesthetic feature extraction and pairwise feature comparison. To alleviate the ambiguity and subjectivity, we train PAC-Net to learn the relative aesthetic ranks of two images by employing a novel loss function, called aesthetic-adaptive cross entropy loss. Then, we develop simple schemes for using PAC-Net in the tasks of aesthetic ranking and aesthetic classification, respectively. Experimental results demonstrate that PAC-Net achieves the state-of-the-art performances in both the ranking and classification applications.
关键词： convolutional neural networks,pairwise comparison,aesthetic ranking,Image aesthetic assessment
更新于2025-09-23 15:23:52
-
Advanced methods for the optical quality assurance of silicon sensors
摘要： We describe a setup for optical quality assurance of silicon microstrip sensors. Pattern recognition algorithms were developed to analyze microscopic scans of the sensors for defects. It is shown that the software has a recognition and classification rate of >90% for defects like scratches, shorts, broken metal lines etc. We have demonstrated that advanced image processing based on neural network techniques is able to further improve the recognition and defect classification rate.
关键词： video microscope,silicon sensors,neural networks,optical quality assurance
更新于2025-09-23 15:23:52