Search

Image segmentation is a crucial task in computer vision, which divides a digital image into multiple segments and objects. In agriculture, image segmentation is extensively used for crop and soil monitoring, predicting the best times to sow, fertilize, and harvest, estimating crop yield, and detecting plant diseases. However, image segmentation faces difficulties in agriculture, such as the challenges of disease staging recognition, labeling inconsistency, and changes in plant morphology with the environment. Consequently, we have conducted a comprehensive review of image segmentation techniques based on deep learning, exploring the development and prospects of image segmentation in agriculture. Deep learning-based image segmentation solutions widely used in agriculture are categorized into eight main groups: encoder-decoder structures, multi-scale and pyramid-based methods, dilated convolutional networks, visual attention models, generative adversarial networks, graph neural networks, instance segmentation networks, and transformer-based models. In addition, the applications of image segmentation methods in agriculture are presented, such as plant disease detection, weed identification, crop growth monitoring, crop yield estimation, and counting. Furthermore, a collection of publicly available plant image segmentation datasets has been reviewed, and the evaluation and comparison of performance for image segmentation algorithms have been conducted on benchmark datasets. Finally, there is a discussion of the challenges and future prospects of image segmentation in agriculture. [ABSTRACT FROM AUTHOR]

Purpose: Excessive application of nitrogen (N) fertilizers in croplands induces serious N pollution and environmental consequences. Nitrification is the primary driver of soil acidification, leading to the consumption of soil inorganic carbon (C) and massive carbon dioxide (CO2) emissions from calcareous soils. Previous studies have shown that N fertilizer with nitrification inhibitors (NIs) can reduce CO2 emissions; however, the underlying mechanisms and factors influencing these reductions remain unclear. Materials and methods: We conducted a comprehensive meta-analysis under different experimental types (incubation and field experiments). The effect of NI amendment on soil CO2 emissions was investigated using data from 57 peer-reviewed scientific articles (172 paired observations), and a machine learning model was used to identify the primary factors influencing this process. Results and discussion: The addition of NIs significantly reduced the CO2 emissions by 19.19% and 7.97% under the incubation and field experimental scenarios, respectively. NIs applied to soils with high pH, low soil organic carbon and total nitrogen content, and coarse texture could mitigate soil CO2 emissions under both the experimental types. CO2 emissions can be considerably reduced by using 3,4-dimethyl pyrazole phosphate (DMPP) or NIs amended with urea or by adopting an N subsurface or single-dose application method. Conclusions: NI amendment has a high potential to reduce the CO2 emissions during N fertilization, and a suitable NI product, soil conditions, and fertilization regimes should be considered to further improve its efficacy. [ABSTRACT FROM AUTHOR]

The growth status of mango (Mangifera indica L.) seedlings directly affects the yield and quality of mango, but there are few reports on water management at the growth stage of mango seedlings. The objective of this study was to identify an optimal water-saving irrigation method for mango seedlings. Taking surface drip irrigation as the control (CK), the effects of alternate infiltration irrigation (AII) on the photosynthetic characteristics, growth, and irrigation water use efficiency (IWUE) of mango seedlings were studied. The AII systems used three irrigation levels: 90% (AII90), 70% (AII70), and 50% (AII50) of the irrigation quota of CK. The results indicated that compared with CK, AII70 and AII50 increased the average leaf water use efficiency by 10.54% and 11.77%, respectively. However, compared with CK, AII50 decreased the average stomatal conductance and transpiration rate by 13.89% and 13.76%, respectively. There was a significant linear relationship between dry mass increment and net photosynthetic rate. The dynamic TOPSIS method indicated that AII70 provided optimal comprehensive photosynthetic benefits. The incremental dry mass and IWUE of AII70 were significantly higher than those of the other treatments. Therefore, we recommend applying a moderate water deficit through alternate infiltration irrigation (AII70) for water management in mango seedling cultivation. [ABSTRACT FROM AUTHOR]

Comprehensive regionalization of fine farming could provide scientific evidence for the programming and distribution of coffee farming industry. Comprehensively considering three factors influencing coffee cultivation such as climate, topography, and soil, 11 evaluation indexes for ecological adaptability were chosen. Ecological factors were transformed to 0–1 by fuzzy membership function. The weighing coefficients of different factors were confirmed by Analytic hierarchy process. Ordinary Kriging interpolation and inverse distance-weighted interpolation in ArcGIS were used. Raster calculator was used for overlay analysis. Thus, the evaluation model of the comprehensive regionalization of coffee farming was established. There were four grades of coffee farming area in Yunnan Province, China by natural discontinuity point method breaks, including the most suitable, the suitable, the less suitable, and the not suitable. Results showed that the lowest average temperature in the coldest month and the altitude had the largest influence on coffee adaptability. Areas of four grades of Coffea arabica accounted for 16.28%, 23.19%, 28.71%, and 31.82% of the total area, respectively, wherein the most suitable area and the suitable area were mainly distributed in areas of Baoshan, Dehong, Puer, Lincang, Yuxi, Xishuangbanna, Honghe, Wenshan, etc. Areas of four grades of Coffea canephora accounted for 2.21%, 21.70%, 36.79%, and 39.30% of the total area, respectively, wherein the most suitable area and the suitable area were mainly distributed in areas of Xishuangbanna, Honghe, Puer, Dehong, Lincang, Wenshan, etc. Results could provide theoretical evidence of spatial planning and risk management of coffee production in Yunnan Province, China. [ABSTRACT FROM AUTHOR]

This paper analyzed the drought situation in southwestern China with the standardized precipitation index (SPI) and the standardized precipitation–evapotranspiration index (SPEI) on multiple time scales. The spatial distribution, seasonal variation and features of drought in southwestern China were examined separately by drought frequency, probability of non-exceedance and runs theory. The results showed that the precipitation sequence will change suddenly near 2073 and an obvious, long-term drought will occur after 2073 due to the decline in precipitation. The spatial distribution of drought frequency indicated that the drought frequency will grow obviously in 2071–2100, and the growth rate will increase from the peripheral to the center of the region. The probability of non-exceedance for SPEI-1 less than − 1.0 will be 14.6% in the spring of 2011–2040, grow in 2041–2070 and reach 25.9% in 2071–2100, while that of SPEI in winter will gradually increase in all periods. When the SPEI-12 is smaller than − 1.0, the non-exceedance probability of the index in all seasons of 1981–2100 was and will be growing gradually, indicating that climate change has and will cause severe droughts in spring and winter from 1981 through 2100, especially the medium- and long-term drought. Through the analysis of the SPI and SPEI time sequences by runs theory, it is observed that the drought severity predicted by SPI-12 was and will be increasing over time, and the long-term drought will become more serious. On all time scales, if the SPEI is smaller than − 1.0, the drought duration will gradually increase from 1981–2010 to 2071–2100, indicating the drought in the study region will worsen due to climate change. [ABSTRACT FROM AUTHOR]

The determination of threshold and the construction of thresholding function would directly affect the signal denoising quality in wavelet transform denoising techniques. However, some deficiencies exist in the conventional methods, such as fixed threshold value and the inflexible thresholding functions. To overcome the defects of the traditional wavelet thresholding techniques, a modified particle swarm optimization (MPSO) algorithm-based parametric wavelet thresholding approach is proposed for signal denoising. Firstly, a kind of parametric wavelet thresholding function construction method is proposed on the basis of conventional thresholding functions. With mathematical derivation, the properties of the constructed function are proved. Three dynamic adjustment strategies are then employed to modify the PSO algorithm. The mean square error (MSE) between the original signal and the reconstructed signal is minimized by the MPSO algorithm. Finally, the performances of the proposed approach and the existing methods are simulated by denoising four benchmark signals with different noise levels. The simulation results show that the proposed MPSO-based parametric wavelet thresholding can obtain lower MSE, higher signal-to-noise ratio, and noise suppression ratio compared to the other algorithms. Besides, the denoising visual results also indicate the superiority of the proposed approach in terms of the signal denoising capability. [ABSTRACT FROM AUTHOR]

In health monitoring of long-span structures, proper arrangement of sensors is a key point because of the need to acquire effective structural health information with limited testing resources. This study proposes a novel approach called dual-structure coding and mutation particle swarm optimization (DSC-MPSO) algorithm for the sensor placement. The cumulative effective modal mass participation factor is firstly derived to select the main contributions modes. A novel method combining dual-structure coding with the mutation operator is then utilized to determine the optimal sensors configurations. Finally, the feasibility of the DSC-MPSO algorithm is verified by optimizing the sensors locations for a long-span cable-stayed bridge. The effective independence method, genetic algorithm and standard particle swarm optimization algorithm are taken as contrast experiments. The simulation results show that the proposed algorithm in this paper could improve the convergence speed and precision. Accordingly, the method is effective in solving optimal sensor placement problems. [ABSTRACT FROM AUTHOR]

Modern control systems usually employ digital technology for controller implementation. The dynamics of the systems are naturally continuous, while control inputs are usually discrete when digital control is utilized. This paper deals with the finite-time stabilization problem of switched systems with sampled-data state feedback under asynchronous switching. The asynchronous switching idea originates from the fact that switching instants of the controllers lag behind or exceed those of subsystems. The attention is focused on designing an asynchronously switched sampled-data controller that guarantees the finite-time stability of the dynamic system. Especially, we consider the case that the switching time and sampling time are not uniform when the system is working. On the basis of finite-time stability theory and multiply Lyapunov functions approach, a finite-time stability condition related to dwell time and sampling period is established. Then, an asynchronously switched sampled-data controller is designed, and the corresponding switching law is also derived to guarantee the considered system to be finite-time stable. Two numerical examples are provided to show the effectiveness of the developed results. [ABSTRACT FROM AUTHOR]

A huge amount of information and identification accuracy in large civil engineering structural damage identification has not been addressed yet. To efficiently solve this problem, a new damage identification method based on rough set and integrated neural network is first proposed. In brief, rough set was used to reduce attributes so as to decrease spatial dimensions of data and extract effective features. And then the reduced attributes will be put into the sub-neural network. The sub-neural network can give the preliminary diagnosis from different aspects of damage. The decision fusion network will give the final damage identification results. The identification examples show that this method can simplify the redundant information to reduce the neural network model, making full use of the range of information to effectively improve the accuracy of structural damage identification. [ABSTRACT FROM AUTHOR]

The role of high current stressing during growth of the P-rich phase at the electroless Ni/Sn interface was examined by transmission electron microscopy. Prior to current stressing, two layers of Ni12P5, columnar Ni12P5 and noncolumnar Ni12P5, were formed after soldering. Upon electric stressing, the two layers of P-rich phase showed opposite growth patterns at the two opposing electrode interfaces. At the cathode, columnar growth of the P-rich phase was greatly enhanced while growth of the noncolumnar layer was inhibited. By contrast, the opposite was found at the anode where the current stressing promoted the noncolumnar growth but suppressed the growth of the columnar layer. Such a strong polarity effect resulted from directional electromigration of the key reaction species, nickel, to and from the interfacial reaction fronts. As a result of the difference in reaction mechanism, overall growth of the P-rich phase was much faster at the cathode during current stressing. [ABSTRACT FROM AUTHOR]

Rice is one of the major food crops, and the study of suitable planting areas for rice plays an important role in improving rice yield and optimizing the production layout. This study used Maximum Entropy (MaxEnt) model to simulate and predict the distribution of suitable rice planting areas in China from 1981 to 2020 by combining the climate, soil, and human activities, analyzed the spatial and temporal changes of suitable rice planting areas in China, and determined the main factors affecting rice planting suitability. The results indicated that the main factors influencing the distribution of suitable planting areas for rice in China were gross domestic product (GDP), population density (Pop), and annual sunshine duration (Sun), with human activities playing a dominant role. The high suitable planting areas of rice were mainly distributed in Hubei, Hunan, Jiangxi, Anhui, Guangdong, southeastern Sichuan and western Guizhou. The total suitable planting areas for rice were 346.00 × 104 km2, 345.66 × 104 km2, 347.01 × 104 km2, and 355.57 × 104 km2 from 1981 to 1990, 1991 to 2000, 2001 to 2010 and 2011 to 2020, respectively. With the passage of time, the area of unsuitable areas for rice gradually decreased, and the area of medium suitable areas increased, with large changes in the area of high- and low-suitable areas. Moreover, due to the transfer of a large number of rural laborers to the cities in recent years, the tension between people and land caused by the population explosion has led to the increasing impact of Pop on rice suitable areas and the relatively weakening of the impact of GDP on rice production interventions. The results can be used to provide scientific evidence for the management of rice cultivation and food production safety, with a view to reducing the impacts of climate change on agricultural production in the context of global climate change. [ABSTRACT FROM AUTHOR]