Search

Currently, complex scene classification strategies are limited to high-definition image scene sets, and low-quality scene sets are overlooked. Although a few studies have focused on artificially noisy images or specific image sets, none have involved actual low-resolution scene images. Therefore, designing classification models around practicality is of paramount importance. To solve the above problems, this paper proposes a two-stage classification optimization algorithm model based on MPSO, thus achieving high-precision classification of low-quality scene images. Firstly, to verify the rationality of the proposed model, three groups of internationally recognized scene datasets were used to conduct comparative experiments with the proposed model and 21 existing methods. It was found that the proposed model performs better, especially in the 15-scene dataset, with 1.54% higher accuracy than the best existing method ResNet-ELM. Secondly, to prove the necessity of the pre-reconstruction stage of the proposed model, the same classification architecture was used to conduct comparative experiments between the proposed reconstruction method and six existing preprocessing methods on the seven self-built low-quality news scene frames. The results show that the proposed model has a higher improvement rate for outdoor scenes. Finally, to test the application potential of the proposed model in outdoor environments, an adaptive test experiment was conducted on the two self-built scene sets affected by lighting and weather. The results indicate that the proposed model is suitable for weather-affected scene classification, with an average accuracy improvement of 1.42%. [ABSTRACT FROM AUTHOR]

Visible-infrared person re-identification aims to solve the matching problem between cross-camera and cross-modal person images. Existing methods strive to perform better cross-modal alignment, but often neglect the critical importance of feature enhancement for achieving better performance. Therefore, we proposed an effective method that combines both modal alignment and feature enhancement. Specifically, we introduced Visible-Infrared Modal Data Augmentation (VIMDA) for visible images to improve modal alignment. Margin MMD-ID Loss was also used to further enhance modal alignment and optimize model convergence. Then, we proposed Multi-Grain Feature Extraction (MGFE) Structure for feature enhancement to further improve recognition performance. Extensive experiments have been carried out on SYSY-MM01 and RegDB. The result indicates that our method outperforms the current state-of-the-art method for visible-infrared person re-identification. Ablation experiments verified the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Aiming at the problem of fast divergence of pure inertial navigation system without correction under the condition of GNSS restricted environment, this paper proposes a multi-mode navigation method with an intelligent virtual sensor based on long short-term memory (LSTM). The training mode, predicting mode, and validation mode for the intelligent virtual sensor are designed. The modes are switching flexibly according to GNSS rejecting situation and the status of the LSTM network of the intelligent virtual sensor. Then the inertial navigation system (INS) is corrected, and the availability of the LSTM network is also maintained. Meanwhile, the fireworks algorithm is adopted to optimize the learning rate and the number of hidden layers of LSTM hyperparameters to improve the estimation performance. The simulation results show that the proposed method can maintain the prediction accuracy of the intelligent virtual sensor online and shorten the training time according to the performance requirements adaptively. Under small sample conditions, the training efficiency and availability ratio of the proposed intelligent virtual sensor are improved significantly more than the neural network (BP) as well as the conventional LSTM network, improving the navigation performance in GNSS restricted environment effectively and efficiently. [ABSTRACT FROM AUTHOR]

An electrochemical sensor enabled by electropolymerization (EP) of β-cyclodextrin on glassy carbon electrode (β-CDP/GCE) is built for the determination of fenitrothion (FNT). The effects of the EP cycles, pH value, and enrichment time on the electrochemical response of FNT were studied. With the optimum conditions, good linear relationships between the current of the reduction peak of the nitroso derivative of FNT and the concentration are obtained in the range of 10–150 and 150–4000 ng/mL, with a detection limit of 6 ng/mL (S/N = 3). β-CDP/GCE also exhibits a satisfactory applicability in cabbage and tap water, with recovery values between 98.43% and 112%. These outstanding results suggest that β-CDP/GCE could be a new effective alternative for the determination of FNT in real samples. [ABSTRACT FROM AUTHOR]

Soil moisture (SM) is a vital climate variable in the interaction process between the Earth's atmosphere and land. However, global soil moisture products from various satellite missions and land surface models are affected by inherently discontinuous observations and coarse spatial resolution, which limits their application at fine spatial scales. To address this problem, this paper integrates three diverse types of datasets from in situ, satellites, and models through Spherical cap harmonic analysis (SCHA) and Helmert variance component estimation (HVCE) to produce 1 km of spatio-temporally continuous SM products with high accuracy. First, this paper eliminates the bias between different datasets and in situ sites and resamples the datasets before data fusion. Then, multi-source SM data fusion is performed based on the SCHA and HVCE methods. Finally, this paper evaluates the fused products from three aspects, including the performance of representative sites under different climate types, the overall performance of validation sites, and the comparison with other products. The results show that the fused products have better performance than other SM products. In the representative sites, the minimal correlation coefficient (R) of the fused products is above 0.85, and the largest root mean square error (RMSE) is below 0.040 m3 m−3. For all validation sites, the R and RMSE of the fused products are 0.889 and 0.036 m3 m−3, respectively, while the R for other products is below 0.75 and the RMSE is above 0.06 m3 m−3. In comparison to other SM products, the fused products exhibit superior performance, generally align more closely with in situ measurements, and possess the ability to accurately and finely capture the spatial and temporal variability of surface SM. [ABSTRACT FROM AUTHOR]

The existing federated learning framework is based on the centralized model coordinator, which still faces serious security challenges such as device differentiated computing power, single point of failure, poor privacy, and lack of Byzantine fault tolerance. In this paper, we propose an asynchronous federated learning system based on permissioned blockchains, using permissioned blockchains as the federated learning server, which is composed of a main-blockchain and multiple sub-blockchains, with each sub-blockchain responsible for partial model parameter updates and the main-blockchain responsible for global model parameter updates. Based on this architecture, a federated learning asynchronous aggregation protocol based on permissioned blockchain is proposed that can effectively alleviate the synchronous federated learning algorithm by integrating the learned model into the blockchain and performing two-order aggregation calculations. Therefore, the overhead of synchronization problems and the reliability of shared data is also guaranteed. We conducted some simulation experiments and the experimental results showed that the proposed architecture could maintain good training performances when dealing with a small number of malicious nodes and differentiated data quality, which has good fault tolerance, and can be applied to edge computing scenarios. [ABSTRACT FROM AUTHOR]

Within the current process of large-scale dairy-cattle breeding, to address the problems of low recognition-accuracy and significant recognition-error associated with existing visual methods, we propose a method for recognizing the feeding behavior of dairy cows, one based on an improved RefineMask instance-segmentation model, and using high-quality detection and segmentation results to realize the recognition of the feeding behavior of dairy cows. Firstly, the input features are better extracted by incorporating the convolutional block attention module into the residual module of the feature extraction network. Secondly, an efficient channel attention module is incorporated into the neck design to achieve efficient integration of feature extraction while avoiding the surge of parameter volume computation. Subsequently, the GIoU loss function is used to increase the area of the prediction frame to optimize the convergence speed of the loss function, thus improving the regression accuracy. Finally, the logic of using mask information to recognize foraging behavior was designed, and the accurate recognition of foraging behavior was achieved according to the segmentation results of the model. We constructed, trained, and tested a cow dataset consisting of 1000 images from 50 different individual cows at peak feeding times. The method's effectiveness, robustness, and accuracy were verified by comparing it with example segmentation algorithms such as MSRCNN, Point_Rend, Cascade_Mask, and ConvNet_V2. The experimental results show that the accuracy of the improved RefineMask algorithm in recognizing the bounding box and accurately determining the segmentation mask is 98.3%, which is higher than that of the benchmark model by 0.7 percentage points; for this, the model parameter count size was 49.96 M, which meets the practical needs of local deployment. In addition, the technologies under study performed well in a variety of scenarios and adapted to various light environments; this research can provide technical support for the analysis of the relationship between cow feeding behavior and feed intake during peak feeding periods. [ABSTRACT FROM AUTHOR]

Livestock's live body dimensions are a pivotal indicator of economic output. Manual measurement is labor-intensive and time-consuming, often eliciting stress responses in the livestock. With the advancement of computer technology, the techniques for livestock live body dimension measurement have progressed rapidly, yielding significant research achievements. This paper presents a comprehensive review of the recent advancements in livestock live body dimension measurement, emphasizing the crucial role of computer-vision-based sensors. The discussion covers three main aspects: sensing data acquisition, sensing data processing, and sensing data analysis. The common techniques and measurement procedures in, and the current research status of, live body dimension measurement are introduced, along with a comparative analysis of their respective merits and drawbacks. Livestock data acquisition is the initial phase of live body dimension measurement, where sensors are employed as data collection equipment to obtain information conducive to precise measurements. Subsequently, the acquired data undergo processing, leveraging techniques such as 3D vision technology, computer graphics, image processing, and deep learning to calculate the measurements accurately. Lastly, this paper addresses the existing challenges within the domain of livestock live body dimension measurement in the livestock industry, highlighting the potential contributions of computer-vision-based sensors. Moreover, it predicts the potential development trends in the realm of high-throughput live body dimension measurement techniques for livestock. [ABSTRACT FROM AUTHOR]

One of the most important ocean water parameters in world ocean observations is temperature. In the application of high-precision ocean sensors, there are often various interferences and random noises. These noises will cause the linearity of the sensor to change, and it is difficult to estimate the statistical characteristics, and the results will deviate from the real temperature. Aiming at the problems in the application, this paper proposes a compound Kalman smoothing filter (CKSF) algorithm based on least square curve fitting. This algorithm first analyzes the system model of the sensor, uses the least square method to fit the theoretical data and eliminate the non-linear factors caused by system itself, then estimates the statistical characteristics of the noise required by modeling, using the wavelet transform method to track the change of noise in real time and to accurately estimate the noise variance. Finally, a compound filtering method including wavelet transform and Kalman smoothing filtering is used as the main denoising algorithm, which is more accurate than a single Kalman filtering result. The algorithm is applied to the temperature measurement process of the ocean temperature sensor. The results show that the accuracy and stability of the sensor are improved. [ABSTRACT FROM AUTHOR]

Drowsiness poses a serious challenge to road safety and various in-cabin sensing technologies have been experimented with to monitor driver alertness. Cameras offer a convenient means for contactless sensing, but they may violate user privacy and require complex algorithms to accommodate user (e.g., sunglasses) and environmental (e.g., lighting conditions) constraints. This paper presents a lightweight convolution neural network that measures eye closure based on eye images captured by a wearable glass prototype, which features a hot mirror-based design that allows the camera to be installed on the glass temples. The experimental results showed that the wearable glass prototype, with the neural network in its core, was highly effective in detecting eye blinks. The blink rate derived from the glass output was highly consistent with an industrial gold standard EyeLink eye-tracker. As eye blink characteristics are sensitive measures of driver drowsiness, the glass prototype and the lightweight neural network presented in this paper would provide a computationally efficient yet viable solution for real-world applications. [ABSTRACT FROM AUTHOR]

Satellite altimetry can provide long-term water level time series for water bodies lacking hydrological stations. Few studies have evaluated the performance of HY-2C and Sentinel-6 satellites in inland water bodies, as they have operated for less than 1 and 2 years, respectively. This study evaluated the measured water level accuracy of CryoSat-2, HY-2B, HY-2C, ICESat-2, Jason-3, Sentinel-3A, and Sentinel-6 in the Great Lakes by in-situ data of 12 hydrological stations from 1 January 2021 to 1 April 2022. Jason-3 and Sentinel-6 have the lowest mean root-mean-square-error (RMSE) of measured water level, which is 0.07 m. The measured water level of Sentinel-6 satellite shows a high correlation at all passing stations, and the average value of all correlation coefficients (R) is also the highest among all satellites, reaching 0.94. The mean RMSE of ICESat-2 satellite is slightly lower than Jason-3 and Sentinel-6, which is 0.09 m. The stability of the average deviation (bias) of the ICESat-2 is the best, with the maximum bias only 0.07 m larger than the minimum bias. ICESat-2 satellite has an exceptionally high spatial resolution. It is the only satellite among the seven satellites that has retrieved water levels around twelve stations. HY-2C satellite has the highest temporal resolution, with a temporal resolution of 7.5 days at station 9075014 in Huron Lake and an average of 10 days in the Great Lakes region. The results show that the seven altimetry satellites currently in operation have their own advantages and disadvantages, Jason-3 and Sentinel-6 have the highest accuracy, ICESat-2 has higher accuracy and the highest spatial resolution, and HY-2C has the highest temporal resolution, although it is less accurate. In summary, with full consideration of accuracy and space-time resolution, the ICESat-2 satellite can be used as the benchmark to achieve the unification of multi-source data and establish water level time series. [ABSTRACT FROM AUTHOR]

Since technologies in image fusion, image splicing, and target recognition have developed rapidly, as the basis of many image applications, the performance of image registration directly affects subsequent work. In this work, for rich features of satellite-borne optical imagery such as panchromatic and multispectral images, the Harris corner algorithm is combined with the scale invariant feature transform (SIFT) operator for feature point extraction. Our rough matching strategy uses the K-D (K-Dimensional) tree combined with the BBF (Best Bin First) method, and the similarity measure is the nearest neighbor/the second-nearest neighbor ratio. Finally, a triangle-area representation (TAR) algorithm is utilized to eliminate false matches in order to ensure registration accuracy. The performance of the proposed algorithm is compared with existing popular algorithms. The experimental results indicate that for visible light and multi-spectral satellite remote sensing images of different sizes and different sources, the proposed algorithm in this work is excellent in accuracy and efficiency. [ABSTRACT FROM AUTHOR]

Image registration is an important basis of image processing, which is of great significance in image mosaicking, target recognition, and change detection. Aiming at the automatic registration problem of multi-angle optical images for ground scenes, a registration method combining point features and line features to register images is proposed. Firstly, the LSD (Line Segment Detector) algorithm is used to extract line features of images. The obtained line segments whose length are less than a given threshold are eliminated by a visual significant algorithm. Then, an affine transform model obtained by estimating a Gaussian mixture model (GMM) is applied to the image to be matched. Lastly, Harris point features are utilized in fine matching to overcome shortages of methods based on line features. In experiments, the proposed algorithm is compared with popular feature-based registration algorithms. The results indicate that the proposed algorithm in this work has obvious advantages in terms of registration accuracy and reliability for optical images acquired at different angles. [ABSTRACT FROM AUTHOR]

Process monitoring of woven fabric thermoplastic composite is crucial to enhance the quality of composite products. In this work, a new fiber Bragg grating based technique was proposed to achieve hybrid temperature and stress monitoring according to the changes of wavelength and reflectivity, respectively. The sensor head consisting of a pre-annealed fiber Bragg grating and a steel capillary was properly designed to overcome the challenge of high forming temperatures up to 332 °C, complex woven structure, and high forming pressure of 2 MPa, which hinder the use of the conventional fiber Bragg grating sensor during the forming process. The forming temperature changes of thermoplastic composite in the heating, dwelling, and cooling phases can be precisely measured by the proposed sensor head after using a curve-reconstruction algorithm based on cubic polynomial fitting. The measured difference from the reference thermocouple is 2.92 °C, averaged from three sets of repeated experiments. Meanwhile, the change of the residual stresses of the composite can be illustrated by using the micro-bending-caused optical power loss of the fiber pigtail commencing at the glass-transition temperature in the cooling phase. The decrease of grating reflectivity that was equivalent to the optical loss was discussed by comparing to strain change detected by strain gauges and a calculated theoretical curve. These results are beneficial for developing an advanced in situ monitoring technique and understanding the forming process of the woven fabric thermoplastic composite. [ABSTRACT FROM AUTHOR]