Search

Dictionary learning has become a research hotspot. How to construct a robust dictionary is a key issue. In face recognition problem, differences in expressions, postures, and lighting conditions are key factors that affect the accuracy. Therefore, images of the same face can be very different in different situations. In real-world scenario, the samples of each face are very limited, which make it hard for the network to generalize well. Therefore, To solve the problem mentioned above, this paper proposes a method to construct a robust dictionary. In the method, virtual samples are generated to appropriately reflect the diversity of the face images, and based on this, two dictionaries are constructed and a corresponding fusion classification scheme is designed. The main advantages of this method are as follows: firstly, the simultaneous use of virtual samples and original samples can better reflect the facial appearance of each morphology, and the dictionaries obtained help to improve the robustness of the dictionary learning algorithm. Secondly, the proposed fusion classification scheme can give full play to the performance of the double dictionary learning algorithm. The results of out experiments show that the proposed algorithm is superior to some existing algorithms.

Hyperspectral remote sensing demonstrates a great potential for identifying crops. Orbita hyperspectral image satellite (OHS) is a new hyperspectral satellite in orbit with the highest spectral and...

The outbreak of large-scale desert locust plague in 2020 has attracted wide attention in the world and caused serious damage to food security and livelihood of African and Asian people. Remote sensing techniques can provide indirect feedback on locust plagues, facilitating quick, and real-time monitoring of the occurrence and development of locusts, which is of great significance for ensuring national and regional food security and stability. The hidden Markov model (HMM) is a classic machine learning model that has been widely applied in the fields of time-series data mining. In this study, we aim to predict the severity of locust plague in croplands using the time-series dynamic change features extracted from remote sensing data via HMM. In addition, we assess the damages on the croplands using change detection methods by comparing the crop spectrum before and after the locust plague from two-phase (Feburary 23 and March 7, 2020) hyperspectral images covering substudy area (northern Narok, Kenya). Evaluated by the ground truth data, the OA of predicted results of the plague severity in April, May, June, and July are 0.78, 0.71, 0.74, and 0.72, respectively. The land cover classification OA of the substudy area of the two-phase images are 97.45 and 96.14. Our study demonstrates the validity of the HMM-based method using the remote sensing time-series data to predict locust plague and evaluate its damage. The results of the cropland change detection suggest that the damage of locusts can be quantitatively evaluated using hyperspectral images.

Global or quasi-global digital elevation model (DEM) datasets provide three-dimensional information on terrain surface, and they have been extremely useful in geoscience research and applications. However, the wide application of DEMs is constrained by differences in the means of observation and processing, and in the resolution of global public DEM datasets. An adaptive regularization variation model based on sparse representation is proposed to generate a high-quality DEM by fusing multi-source DEMs. First, since the sparse representation method has a powerful capability to reconstruct information based on a small amount of information, prior terrain information is extracted from the 90-m TanDEM-X DEM (TDM90) with unprecedented global accuracy using a so-called sparse representation. In this step, an intermediate DEM (termed STDM30) is first extracted from TDM90 that preserves maximum terrain details, thereby preventing the degradation of the DEM accuracy induced by resampling. Then, the designed regularization framework based on terrain slope can constrain the DEM spatial information during fusing multiple datasets. STDM30 is combined with the ALOS Global Digital Surface Model “ALOS World 3D 30 m” (AW3D30) and the 1 arc-second Shuttle Radar Topography Mission Digital Elevation Model (SRTM1) through the designed adaptive regularization variation model to generate a high-accuracy DEM product with a resolution of 30 m. The results of the proposed method were verified by a model-to-model comparison in South Dakota as well as by validation against GPS benchmarks in Southern California. The RMSE, MAE, and SD of the fused DEM are all lower than those of the existing public DEMs, especially in terms of removing topographic noise and refining terrain details. The GPS validation showed that the fused DEM has an RMSE of 3.04 m, with the highest absolute accuracy among the four studied DEMs, and its errors are almost equal to the normal distribution. These experimental results confirm that the multi-scale and multi-source DEM fusion strategy combining sparse representation and an adaptive regularization variation model can utilize existing public datasets and effectively improve the quality of global DEM products.

The semantic segmentation of remote sensing (RS) image is a hot research field. With the development of deep learning, the semantic segmentation based on a full convolution neural network greatly improves the segmentation accuracy. The amount of information on the RS image is very large, but the sample size is extremely uneven. Therefore, even the common network can segment RS images to a certain extent, but the segmentation accuracy can still be greatly improved. The common neural network deepens the network to improve the classification accuracy, but it has a lot of loss to the target spatial features and scale features, and the existing common feature fusion methods can only solve some problems. A segmentation network is built to solve the above problems very well. The network employs the InceptionV-4 network as the backbone and improves it. We modify the network structure and introduce the changed Atrous Spatial Pyramid Pooling module to extract the multi-scale features of the target from different training stages. Without losing the depth of the network, using Inception blocks to strengthen the width of the network can obtain more abstract features. At the same time, the backbone network is used for semantic fusion of the context, it can retain more spatial features, then an effective decoder network is designed. Finally, evaluate our model on the ISPRS 2D Semantic Labeling Contest Potsdam and Inria Aerial Image Labeling Dataset. The results show that the network has very superior performance, reaching 89.62% IOU score and 94.49% F1 score on the Potsdam dataset, and the IOU score on the Inria dataset has been greatly improved.

License plate detection and recognition are still important and challenging tasks in natural scenes. At present, most methods have favorable effect on license plate recognition under restrictive conditions, and most of such license plates are shot under good angle and light conditions. However, for license plates under non-restrictive conditions, such as dark, bright, rotated conditions etc. from the Chinese City Parking Dataset (CCPD), the performance of some methods of license plate recognition will be significantly reduced. In order to improve the accuracy of license plate recognition under unrestricted conditions, a robust license plate recognition model is proposed in this paper, which mainly includes license plate feature extraction, license plate character localization, and feature extraction of characters. First of all, the model can activate the regional features of characters and fully extract the character features of license plates. Then locate each license plate character through Bi-LSTM combined with the context location information of license plates. Finally, 1D-Attention is adopted to enhance useful character features after Bi-LSTM positioning, and reduce useless character features to realize effective acquisition of character features of license plates. A large number of experimental results demonstrate that the proposed algorithm has good performance under unrestricted conditions, which proves the effectiveness and robustness of the model. In CCPD-Base, CCPD-DB, CCPD-FN, CCPD-Tilt, CCPD-Weather, CCPD-Challenge and other sub-datasets, the recognition rates reach 99.3%, 98.5%, 98.6%, 96.4%, 99.3% and 86.6% respectively.

Pedestrian detection is a hot and difficult topic in the computer vision field. The Histograms of Oriented Gradients (HOG) feature, because of its high performance in accuracy, is widely used in pedestrian detection. Nonetheless, its information description capacity needs further improvement. so, I-HOG (Improved HOG) was proposed. I-HOG has two major improvements. First, I-HOG enhances the description of edge features. Through the different scales for the block histograms of a set of correlation graphs, makes the correlation between characteristic information. Second, I-HOG using multi-scale feature extraction methods, include wider edge feature description information, make up for the deficiencies of the HOG feature, because HOG features are only extracted in fixed block size, The experimental results show that in the INRIA database, using I-HOG, detection rate increased by 5.4% and 4.3% respectively, combined with the feature of CSS after detection rate increased by 2.8% and 4.0% respectively compared to the HOG.

The hexagonal boron nitride (h-BN) nanosheets modified by silane coupling agent (KH560) were doped into acrylic acid coating on the surface of galvanized steel to improve its corrosion resistance. H-BN nanosheets modified by KH560 were prepared and characterised by scanning electron microscopy, transmission electron microscopy, atomic force microscopy, X-ray diffraction, and Fourier-transform infrared spectroscopy. The corrosion resistance of the acrylic acid coatings was measured by electrochemical testing. The results show that the corrosion current density of the coating with modified h-BN nanosheets was reduced from 2.2 &times, 10&minus, 5 A/cm2 to 2.3 &times, 7 A/cm2 compared with the acrylic acid coating. The impedance of the composite coating with modified h-BN is 4435 &Omega, &middot, cm2, higher than the BNNS coating (2500 &Omega, cm2) and the acrylic acid coating (1500 &Omega, cm2). This is due to the physical barrier and electrical insulation properties of the hexagonal boron nitride (h-BN) nanosheets.

Based on the big-data collected from Space-Air-Ground, i.e. Space means satellite, Air means helicopter, the key technologies of novel ITS (Intelligent Transportation System) are investigated, including data acquisition sensor, dynamic data transmission, massive data storage, multi-source data fusion, massive data mining and analysis, etc. On this basis, the cloud computing platform of novel ITS is designed, including Space-Air-Ground bigdata acquisition & transmission subsystem, cloud computing platform, intelligent transportation application & service subsystem. With the help of the data visualization, data prediction, and decision making, the complete traffic big-data set including people (passenger, driver), vehicle, and road traffic environment, can create their core addedvalues. The applications of novel ITS include: providing transportation data services for traffic enterprise and business users, such as customized mining, and specific industry analysis; providing accurate transportation information services for the citizen; providing business model for all levels of users, such as data visualization and customized services.

Due to the influence of germs and viruses, plants often show various symptoms of diseases and insect pests during the growth process, which leads to a large economic loss of fruit farmers. It also brings a certain economic loss to our society, so prevent earlier and advise growers about plant diseases and insect pests have important value and significance. In this case, this paper proposes a detection method which is based on the combination of HOG, LBP and CSS features with Support Vector Machine (SVM) classifier. This method extracts the histogram of oriented gradients, texture, and color self-similar features of potato leaves, and then training samples with SVM classifier to detect late blight as early as possible in the early stages of potato growth. In addition, this paper proposes a method to increase virtual samples, that is, generating symmetrical samples according to the original samples. Due to the limitation of the number of collected samples, increasing symmetrical samples can expand the diversity of samples. The results show that this method can obtain a detection rate of 92.7%, and has better detection and recognition performance in practical application.

A two-dimensional steady-state mathematical model is developed to study the heat and water vapor transport in a run-around heat and moisture exchanger coupled with a lithium bromide solution for air-to-air exchanger applications. A finite difference method is employed to solve the governing equations of the heat and moisture exchange, which gives the outlet air properties and effectiveness for selected operating conditions for each cross-flow exchanger. Using algorithms for the HVAC supply and exhaust exchangers coupled with a run-around liquid loop, the overall effectiveness of the run-around energy recovery system is shown to be dependent on the flow rate of both the pumped fluid and each airflow, the size and design of each exchanger, and the inlet operating conditions. It is shown that an overall effectiveness of 70% can be achieved when the run-around exchanger sizes and operating conditions are correctly chosen.

Based on the Big Data collected from Space-Air-Ground, novel key technologies of intelligent transportation system are investigated, like sensor data acquisition, dynamic data transmission, mass data storage, multisource data fusion, data mining and analysis, etc. On this basis, a novel intelligent transportation cloud platform is designed, including Space-Air-Ground Big Data acquisition and transmission subsystem, cloud computing supporting platform, intelligent transportation application and service subsystem. With the help of the data visualization, data prediction, and decision-making, the whole traffic Big Data including people (passenger, driver), vehicle, and road traffic environment, can create the core added-values. The applications are: providing transportation data services for traffic enterprise users and business users, such as customized mining, and specific industry analysis; providing accurate transportation information services for the public; providing business model for all levels of users, such as data visualization and customized services.