Search

The traditional behavior recognition model has the disadvantage that it can't get the internal relationship between similar behaviors, such as smoking, pen, chin and the clamped objects, which limits the actual landing of such fine and complex behaviors as smoking recognition. To solve these problems, this paper puts forward the heterogeneous algorithm HMMA-NET (Heterogeneous multi-task smoking behavior recognition model combined with Attention), which consists of two modules: behavior prior and local detection, aiming at establishing the relationship between behavior and behavior objects. CNN combined with channel attention mechanism is used in both behavior prior module and local detection module. The former uses sign language semantic features to complete the primary prior of behavior according to the obtained behavior affinity vector field, while the latter designs network optimization such as fast Edgebox to obtain candidate areas, so as to transfer component information and achieve the goal of fast fine-grained detection. Finally, the two modules use SaaS mode to complete association recognition. Experiment shows that the algorithm can recognize complex actions effectively, and its accuracy is still equal to or even better than that of a single model, in which the accuracy of detecting smoking behavior scenes is 96.10%, and the false detection rate is 3.6%. The algorithm has been commercialized and applied to the actual monitoring of petrochemical scenes. The running results show that the algorithm can maintain good real-time performance and generalization ability. [ABSTRACT FROM AUTHOR]

Aeromagnetic compensation is important in aeromagnetic detection. Traditionally, a calibration flight would be implemented to estimate the interference model. However, the magnetic interference is not constant during the mission flight, which makes the interference model constructed in the calibration flight inapplicable in the mission. To solve that problem, this letter proposes an evolutionary aeromagnetic compensation method based on the Woodbury equation. With this method, the magnetic interference model parameters can be evolved during the mission flight on basis of the previous interference model and the updated acquired data. To evaluate the performance of the evolutionary method, both simulation and the experiment were conducted. The results indicate that the proposed method can effectively reduce the influence of changing magnetic interference. In the experiment, the standard deviation (STD) of measured data before compensation in the last mission flight is 1.2070 nT, the traditional compensation can reduce the STD to 0.0863 nT, while our evolutionary aeromagnetic compensation method can reduce STD to 0.0450 nT. The results show that increasing changes in magnetic interference will make evolutionary aeromagnetic compensation vital for isolating signals created by geologic features from signals created by the aircraft. [ABSTRACT FROM AUTHOR]

Magnetic interferential compensation plays a vital role in geomagnetic vector measurement applications. Traditional compensation accounts for only the permanent interferences, induced field interferences, and eddy-current interferences. However, nonlinear magnetic interferences are found, which also have a great impact on measurement, and it cannot be fully characterized by a linear compensation model. This paper proposes a new compensation method based on a back propagation neural network, which can reduce the influence of the linear model on compensation accuracy due to its good nonlinear mapping capabilities. The high-quality network training requires representative datasets, yet it is a common problem in the engineering field. To provide adequate data, this paper adopts a 3D Helmholtz coil to restore the magnetic signal of a geomagnetic vector measurement system. A 3D Helmholtz coil is more flexible and practical than the geomagnetic vector measurement system itself when generating abundant data under different postures and applications. Simulations and experiments are both conducted to prove the superiority of the proposed method. According to the experiment, the proposed method can reduce the root mean square errors of north, east, and vertical components and the total intensity from 73.25, 68.54, 70.45, and 101.77 nT to 23.35, 23.58, 27.42, and 29.72 nT, respectively, compared with the traditional method. [ABSTRACT FROM AUTHOR]

Owing to the ferromagnetism interferential field, the component compensation of magnetic field distortion is vital for a geomagnetic field vector measurement system. However, magnetic interferences, such as eddy-current field, have not been considered in the traditional component compensation methods due to the reference component magnetic field, and its variation with time is difficult to obtain. In this article, a new compensation method using a 3-D Helmholtz coil is proposed, in which the eddy-current field related to the reference component magnetic field and its variation with time are integrated into a compensation model. In addition, the rotation equipment is not necessarily needed to construct the equations of an error model, and the 3-D Helmholtz coil is used to generate any required compensation data to estimate the compensation model error parameters, which makes the compensation procedure more simple and convenient. In order to verify the effectiveness of the proposed method, the experiment is conducted, and the results demonstrate that the proposed method contributes to the accuracy improvement of the geomagnetic vector measurement system. After compensation, the rms errors of North, East, Vertical components, and total intensity are reduced from 3648.3, 2986.2, 3496.2, and 3894.1 to 38.72, 29.58, 35.42, and 42.82 nT, respectively. [ABSTRACT FROM AUTHOR]

Simple Summary: Since animal domestication, breeders have been selecting candidates for breeding based on phenotypic performance. Estimating breeding values through the best linear unbiased prediction method represents a revolutionary shift in animal breeding. On this basis, selection and mating are utilized to improve the production level of animals. The application of genomic selection has once again revolutionized animal breeding methods. However, although this kind of truncated selection based on breeding values can significantly improve genetic gain, the genetic relationship between individuals with a high breeding value is usually closed, and the probability of being co-selected is greater, which will lead to a rapid increase in the rate of inbreeding in the population. Reduced genetic variation is not conducive to long-term sustainable breeding, so a trade-off between genetic gain and inbreeding is required. Genomic mating is the use of candidate individuals' genomic information to implement optimized breeding and mating, which can effectively control the rate of inbreeding in the population and achieve long-term and sustainable genetic gain. It is more suitable for modern animal breeding, especially for conservation and genetic improvement of local domestic animal breeds. Selection is a continuous process that can influence the distribution of target traits in a population. From the perspective of breeding, elite individuals are selected for breeding, which is called truncated selection. With the introduction and application of the best linear unbiased prediction (BLUP) method, breeders began to use pedigree-based estimated breeding values (EBV) to select candidates for the genetic improvement of complex traits. Although truncated selection based on EBV can significantly improve the genetic progress, the genetic relationships between individuals with a high breeding value are usually closed, and the probability of being co-selected is greater, which will lead to a rapid increase in the level of inbreeding in the population. Reduced genetic variation is not conducive to long-term sustainable breeding, so a trade-off between genetic progress and inbreeding is required. As livestock and poultry breeding enters the genomic era, using genomic information to obtain optimal mating plans has formally been proposed by Akdemir et al., a method called genomic mating (GM). GM is more accurate and reliable than using pedigree information. Moreover, it can effectively control the inbreeding level of the population and achieve long-term and sustainable genetic gain. Hence, GM is more suitable for modern animal breeding, especially for local livestock and poultry breed conservation and genetic improvement. This review mainly summarized the principle of genomic mating, the methodology and usage of genomic mating, and the progress of its application in livestock and poultry. [ABSTRACT FROM AUTHOR]

The current development of blockchain, technically speaking, still faces many key problems such as efficiency and scalability issues, and any distributed system faces the problem of how to balance consistency, availability, and fault tolerance need to be solved urgently. The advantage of blockchain is decentralization, and the most important thing in a decentralized system is how to make nodes reach a consensus quickly. This research mainly discusses the blockchain and K-means algorithm for edge AI computing. The natural pan-central distributed trustworthiness of blockchain provides new ideas for designing the framework and paradigm of edge AI computing. In edge AI computing, multiple devices running AI algorithms are scattered across the edge network. When it comes to decentralized management, blockchain is the underlying technology of the Bitcoin system. Due to its characteristics of immutability, traceability, and consensus mechanism of transaction data storage, it has recently received extensive attention. Blockchain technology is essentially a public ledger. This is done by recording data related to trust management to this ledger. To collaboratively complete artificial intelligence computing tasks or jointly make intelligent group decisions, frequent communication is required between these devices. By integrating idle computing resources in an area, a distributed edge computing platform is formed. Users obtain benefits by sharing their computing resources, and nodes in need complete computing tasks through the shared platform. In view of the identity security problems faced in the sharing process, this article introduces blockchain technology to realize the trust between users. All participants must register a secure identity in the blockchain network and conduct transactions in this security system. A K-means algorithm suitable for edge environments is proposed to identify different degradation stages of equipment operation reflected by multiple types of data. Based on the prediction of the fault state for a single type of data, the algorithm uses the historical data of multiple types of data together with the prediction data to predict the fault stage. During the research process, the average optimization energy consumption of K-means algorithm is 14.6% lower than that of GA. On the basis of designing a resource allocation scheme based on blockchain, the problem of how the participants can realize reliable resource use according to the recorded data on the chain is studied. The article implements the verification of the legality of the use of blockchain resources. In addition, a control node is introduced to master the global real-time information of the network to provide data support for the user's choice. [ABSTRACT FROM AUTHOR]

Here we used two kinds of chips data from 5 pig breeds, Chinese Duroc (DD), Landrace (LL), Yorkshire (YY), Liangshan (LS), and Qingyu pigs (QY) in China to identify genes which show evidence of selection during domestication. Four breed pairs, LS-YY, QY-YY, DD-YY, and LL-YY pair, were performed to detect selection signatures using the Fst method. Then we identified a list of genes that played key roles in domestication and artificial selection. For example, the PTPRM gene was shared in LS-YY, QY-YY, and DD-YY pairs and it regulates a variety of cellular processes including cell growth, differentiation as signaling molecules. The HACD3 gene was shared in QY-YY and DD-YY pairs, and the HACD3 protein is involved in the production of very long-chain fatty acids of different chain lengths. Besides, the MYH11 gene that related to muscle contraction was found in LS-YY and LL-YY pair. These results suggested that genes related to immunity, disease resistance, and metabolism were subjected to strong selection pressure in Chinese domestic pigs in the progress of domestication and evolution; however, genes related to appearance, production performance, and reproduction were undergone strong artificial selection in commercial pig breeds. [ABSTRACT FROM AUTHOR]

Socially affected traits in pigs are controlled by direct genetic effects and social genetic effects, which can make elucidation of their genetic architecture challenging. We evaluated the genetic basis of direct genetic effects and social genetic effects by combining single-locus and haplotype-based GWAS on imputed whole-genome sequences. Nineteen SNPs and 25 haplotype loci are identified for direct genetic effects on four traits: average daily feed intake, average daily gain, days to 100 kg and time in feeder per day. Nineteen SNPs and 11 haplotype loci are identified for social genetic effects on average daily feed intake, average daily gain, days to 100 kg and feeding speed. Two significant SNPs from single-locus GWAS (SSC6:18,635,874 and SSC6:18,635,895) are shared by a significant haplotype locus with haplotype alleles 'GGG' for both direct genetic effects and social genetic effects in average daily feed intake. A candidate gene, MT3, which is involved in growth, nervous, and immune processes, is identified. We demonstrate the genetic differences between direct genetic effects and social genetic effects and provide an anchor for investigating the genetic architecture underlying direct genetic effects and social genetic effects on socially affected traits in pigs. Pingxian Wu and Kai Wang et al. perform single-locus and haplotype-based GWAS in Yorkshire pigs to evaluate genetic factors involved in socially affected traits, like feed intake and weight gain. These results indicate several genetic factors that may ultimately be targeted to improve livestock breeding. [ABSTRACT FROM AUTHOR]

A thermal barrier coating (TBC), which is composed of a top coating (TC) and bond coating (BC), can keep a turbine engine working in high temperature. The TC is an insulated ceramic layer, and the BC is a conductive layer between the TC and engine blade. Owing to poor working conditions, some failures such as sintering, thinning of coating thickness, and oxide layer initiation will occur in the TBC. Once any part of the TBC fails, it will seriously threaten the safety of the aircraft. The quantitative detection of TBC parameters is realized with the electromagnetic/capacitive dual modality sensor in this paper. The measurement grid algorithm is used to inverse the thickness of the TC layer and the conductivity of the BC layer, and an analytical method is proposed to inverse the relative permittivity of the TC layer. According to the experiment, the inversion errors of these parameters are all less than 4%, which can meet the industry needs well. [ABSTRACT FROM AUTHOR]

The objective of this study was to estimate group‐ and breed‐specific genetic parameters for reproductive traits in Chinese Duroc, Landrace, and Yorkshire populations. Records for reproductive traits between April 1998 and December 2017 from 92 nucleus pig breeding farms, which were involved in the China Swine Genetic Improvement Program, were analysed. Due to weak genetic connectedness across all farms, connectedness groups consisting of related farms were used. Three, two and four connectedness groups for Duroc, Landrace and Yorkshire were firstly established according to the genetic connectedness rating among farms. For each connectedness group a five‐trait animal model was implemented, and via restricted maximum likelihood procedure the genetic parameters were estimated for five reproductive traits i.e., total number born (TNB), number born alive (NBA), litter weight at farrowing (LWF), farrowing interval (FI) and age at first farrowing (AFF). The average of heritabilities among connectedness groups ranged from.01 (for FI in Yorkshire) to.30 (for AFF in Duroc). Estimates of repeatability for litter traits ranged from.14 to.20 and were consistent for each breed, and for FI, the estimates varied from.01 to.11 across breeds and groups. The estimated genetic correlations among litter traits (i.e., TNB, NBA and LWF) were all significantly high (>.56) and similar across breeds. Averaged genetic correlations over three breeds were −.25, −.27, −.18, −.04, −.10, −.02, and.28 for FI‐TNB, FI‐NBA, FI‐LWF, AFF‐TNB, AFF‐NBA, AFF‐LWF and FI‐AFF, respectively. The standard errors of the estimates were all very low (<0.01) in most situations. Results from this study suggest that selection based on TNB which is currently used in dam line selection index can improve NBA and LWF simultaneously. However, care should be taken on FI and AFF as they are both greatly influenced by non‐genetic factors such as management and measurement. [ABSTRACT FROM AUTHOR]

The abundance of single nucleotide polymorphisms (SNPs) makes the haplotype-based method instead of single-maker-oriented method the main approach to association studies on QTL mapping. The key problem in haploptype-based method is how to reconstruct haplotypes from genotype data. Directly assaying haplotypes in diploid individuals by experimental methods is too expensive, therefore the in silico haplotyping-determination methods are the major choice at the present. This paper presents a rapid and reliable algorithm for haplotype reconstruction for tightly linked SNPs in general pedigrees. It is based on six rules and consists of three steps. First, the parental origins of alleles in offspring are assigned conditional on genotypes in parent-offspring trios; second, the redundant haplotypes are eliminated based on the six rules; and finally, the most likely haplotype combinations are chosen via maximum likelihood method. Our method was verified and compared with PEDPHASE by simulated data with different pedigree sizes, numbers of loci, and proportions of missing genotypes. The result shows that our algorithm was superior to PEDPHASE in terms of computing time and accuracy of haplotype estimation. The computing time for 100 runs was 10–15 times less and the accuracy was 4%–10% higher than PEDPHASE. The result also indicates that our method was very robust and was hardly affected by pedigree size, number of loci, and proportion of missing genotypes. [ABSTRACT FROM AUTHOR]

A Chinese Holstein population with daughter design was analyzed using 14 microsatellites covering a map distance of 557 cM on chromosome 6 to fine map QTL for five milk production traits. 26 paternal half-sib families with 2356 daughters were involved. Two different approaches, linear regression approach and variance component approach, were employed, with a one-QTL model and two-QTL model fitted. With a one-QTL model, the linear regression approach revealed a QTL near BMS470 with effects on milk yield, fat yield, protein yield, and fat percentage, and another QTL near BMS2460 for protein percentage. The variance component approach confirmed the results of linear regression approach for the three yield traits, with the exception that the QTL for fat yield was mapped to a different position near BMS1242. The 95% confidence intervals resulted from linear regression, obtained by bootstrapping, were generally large, ranging from 31 to 53 cM, whereas the variance component approach revealed very small confidence intervals, calculated by LOD drop-off method, for the three yield traits, only 4-5 cM. With a two-QTL model, both approaches provided strong evidence for the existence of two QTLs for the three yield traits. Along with the QTLs identified in one-QTL model analyses, the linear regression approach revealed a second QTL near BP7 with effects on all the three yield traits, whereas the variance component approach located the second QTL near ILSS035, BMS470, and BP7 for the three traits, respectively. [ABSTRACT FROM AUTHOR]

Recognition of surface targets has a vital influence on the development of military and civilian applications such as maritime rescue patrols, illegal-vessel screening, and maritime operation monitoring. However, owing to the interference of visual similarity and environmental variations and the lack of high-quality datasets, accurate recognition of surface targets has always been a challenging task. In this paper, we introduce a multi-attention residual model based on deep learning methods, in which channel and spatial attention modules are applied for feature fusion. In addition, we use transfer learning to improve the feature expression capabilities of the model under conditions of limited data. A function based on metric learning is adopted to increase the distance between different classes. Finally, a dataset with eight types of surface targets is established. Comparative experiments on our self-built dataset show that the proposed method focuses more on discriminative regions, avoiding problems like gradient disappearance, and achieves better classification results than B-CNN, RA-CNN, MAMC, and MA-CNN, DFL-CNN. [ABSTRACT FROM AUTHOR]