Search

Water stress has been one of the most serious threat to the crop growth, development and yield quality in agricultural fields. Timely and accurate diagnosis of crop water stress can greatly contribute to the precision irrigation for the crop resilience and yield. In this study, the research object was taken from the summer maize in the typical dryland agricultural area of northwest China. A six-channel multispectral sensor was mounted on a drone to obtain the remote sensing image data of summer maize at the nodulatione and staminate pulling stage in 2022. At the same time, the stomatal conductance and phenotypic parameters of summer maize were also collected. The background was removed by supervised classification. The canopy vegetation index and image texture were obtained using the gray-scale covariance matrix. The sensitive vegetation index, image texture and phenotypic parameters and their combinations were screened out by the Bayesian information criterion and full subset filtering. The summer maize stomatal conductance estimation model was constructed to combine the three types of machine learnings: the extreme learning machine, the random forest, and the back-propagation neural network. The optimal model was mapped to estimate the stomatal conductance. The Pearson correlation coefficient of vegetation index and stomatal conductance were significantly positively correlated, whereas, the canopy reflectance of summer maize was weakly negatively correlated. Different types of image textures at different wavelengths were correlated with the stomatal conductance, and the highest correlation was found in the 550 nm band. The Pearson correlation coefficients between morphological structure phenotypes (plant height, stem thickness and leaf area) and stomatal conductance of summer maize were 0.72, 0.58 and 0.69, respectively, where the three types of phenotypic parameters data were correlated well with stomatal conductance. Vegetation indices with spectral reflectance data were used to assess the overall health and moisture status of the vegetation. Image texture was used to capture the spatial distribution, texture and structural features of crops. Crop phenotypic parameters were then used to reflect the physiological and morphological responses of the crop in a three-dimensional manner, providing visual information about the growth and moisture of the vegetation. The decision coefficients of the crop water stress diagnostic models that constructed from the three information sources increased from 0.728 and 0.750 to 0.841, respectively, compared with the single or two combinations, indicating the great potential to stomatal conductance prediction. The optimal combination of indicators was screened by Bayesian information criterion and full subset screening: DWSI, NDVI, MEA, ENT, plant height and leaf area. The back-propagation neural network model with the three complementary information sources was the optimal model for the water stress diagnosis of summer maize (coefficient of determination of 0.841, root mean square error of 0.043 mol/(m2 ·s), and mean absolute error of 0.034 mol/(m² ·s)). The underestimation of stomatal conductance was significantly improved, compared with the rest models. The inverse map with the optimal model was widely applied to easily and accurately diagnose the crop water stress for the purpose of irrigation strategies and resource allocation. The finding can provide a feasible and accurate diagnosis of water stress in summer maize. [ABSTRACT FROM AUTHOR]

Currently, about 60% of the world’s fish come from China. Our nation is the biggest aquaculture country in the world. However, China's traditional aquaculture farming methods still plays an important role. This will lead to a waste of hunman resources and land resources. In recent years, more than 90% of the fish farming workers prefer to apply RAS (recirulating aquaculture systems). Because RAS (recirulating aquaculture systems) not only create a good rearing conditions and feed utilization for the good growth of the shoal, but also provide a necessary prerequisite for the sustainable development of the aquaculture. Although RAS has a total of more convenience, nevertheless, it still has a lot of problems: the over-feeding events continues to affect the welfare of shoal, the operation of systems also influence the growth and survival of shoal, in addition, the water quality of the system increasing the risk of diseases and leading to extra-load on the mechanical filters, bio-filters and oxygenation equipments. Therefore, the most direct and effective way to solve this problem is to determine the amount of feeding activities according to the appetite of the shoal. In actual operation, manual observation is an effective method. But, on the one hand, this way requires fish-farming workers to increase labor and time costs. On the other hand, the most difficult problem is how to measure feeding activity. This is becauset he reslut of visual observation for the shoal appetite is often hampered by high shoal density and some other unfavorable factors .In recent years, several methods have been developed to enable tracking of several fish in the same tank, but these methods are invasive, potentially disturbing, or need restricting conditions. In order to solve this problem, a novel method, with the help of the original method based on image texture of the Cyprinus carpio specularis, is proposed to evaluate the feeding activity of the shoal. With this method, the feeding activity of Cyprinus carpio specularis is recorded by a camera mounted above the rearing tank. Then, the video is decomposed into a sequence frame image, remove the background interference, extract the foreground object. Next, the texture of shoal is extracted by using gray level cooccurrence matrix. By this way, it can avoid the interference of surface vibration, spray and other adverse factors. At last, we find out the similarity and difference of texture features. To sum up the above arguments, this method can be used for quantifing analysis of the feeding activity of the shoal. To evaluate the reliability of shoal image texture, the other way is put forward to quantify the feeding activity of shoal. Before this research, the diffusion and aggregation indices used by some researchers describe the feeding activity of shoal. The index is calculated by summing the individual area of each group. A group is a single black area created by the outlines of overlapping fish. Using this method, it is easy to observe that when the shoal is spread, the index would be high, otherwise, the index would be low. The correlation coefficient of the above methods is 0.8942. Hence, the feeding activity of fish can be quantified by the image texture of the shoal. This method provides a sensitive, non-invasive, simple and widely applicable tool to quantify the behavior changes associated with the various challenges in aquaculture conditions. [ABSTRACT FROM AUTHOR]

Cotton is one of the most important economic crops in China. It is of great significance to predict the yield of cotton. In this study, the Unmanned Aerial Vehicle (UAV) remote sensing platform was first used to collect some data on the densely planted cotton in Xinjiang, China. Five-point sampling was also utilized in the period of defoliation. A total of five cotton fields were then selected as well. Secondly, each image was divided into multiple sub-images and then enhanced by color space conversion, brightness, noise blurring, flipping and rotating operation. A high-quality dataset was constructed to estimate the cotton yield. Better training was achieved, compared with the original. Eventually, the cotton dataset with the construction site was used to extract the element rate from the cotton images using RCH-UNet (resnet coordinate hardswish UNet). Among them, the UNet was used as a baseline model to construct an improved U-type convolutional neural network. While the IoU (intersection over union), PA (pixel accuracy), and precision were taken as the evaluation metrics. The overall RCH-UNet improved the three metrics by 14.34, 9.85, and 8.68 percentage points, respectively, compared with the original UNet. Specifically, ResNet50 backbone feature extraction network was selected to replace the traditional CBR (convolution batch normalization ReLU) downs sampling structure in UNet; The CA (coordinate attention) mechanism was combined with the UNet; The ReLU activation function in UNet was replaced with the hardswish activation function. The results showed that: 1) The richer semantic information in the ResNet50 was learned through the residual structure, indicating the better feature extraction and expression of the improved model. 2) The CA mechanism was significantly enhanced to learn the detailed features. At the same time, there was an effective reduction in the interference of irrelevant features to the model. 3) The hardswish activation function presented the stronger expression and feature fusion of UNet when performing up-sampling and jump connection. Subsequently, the performance of RCH-UNet was tested with the PSPNet and DeepLabv3 models under the same experimental conditions. The IoU, PA, and precision of RCH-UNet had improved by 9.15, 6.31, and 3.99 percentage points, respectively, compared with the DeepLabv3 model. A prediction model of cotton yield was constructed with the ridge regression from the cotton pixels extracted by the RCH-UNet model. The image texture features were also extracted by GLCM (gray-level co-occurrence matrix). The R² value of the improved model was 0.92, and the average relative error between the predicted and actual yield was 9.254%. The RCH-UNet model was accurately and effectively extracted from the cotton images; Meanwhile, the prediction model of cotton yield was effectively verified using deep learning and image processing. The UAV low-altitude imaging can also be expected to predict the yield of densely planted cotton in Xinjiang, China. [ABSTRACT FROM AUTHOR]

Ultra-high-resolution aerial images obtained from Unmanned Aerial Vehicles (UAVs) have widely been used to extract crop planting information in recent years. However, some high-resolution multispectral or hyperspectral images were usually costly and time-consuming for data processing. Therefore, it is very necessary to effectively use easily accessible and low-cost high-resolution RGB images, particularly to eliminate the common noises (e.g., shadows and bare land) for a better extraction accuracy of crop planting. In this study, a high-precision extraction method was proposed to obtain the maize planting information using 1.8 cm resolution UAV aerial orthophotos (i.e., RGB images). The experimental maize farm was located in Southeast Africa, where images were taken at noon during the maize growing season. The classification features were also selected from the aspects of the spectrum, color space, and image texture. Then, five types of classification were selected to extract maize planting information, including Bayes, K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF) algorithms). Firstly, an object-oriented interpreting platform (eCognition9.0 software) was selected to calculate the space transform of Hue, Saturation, and Intensity (HSI) color, eight types of RGB texture, and five vegetation indices, including the normalized green-red difference index, red-green ratio index, vegetation color verification index, visible-band difference vegetation index, and excess green vegetation index. Then, three types of feature space were constructed: 1) The first feature space was composed of three sub-feature spaces, i.e., vegetation indices, HSI color space features, and RGB image texture features; 2) The second feature space was composed of four sub-feature spaces, where three types of features were combined (i.e., vegetation indices, HSI color space, and RGB image texture) in pairs or total; 3) The third feature space was composed of the most optimal factors, where the dimension reduction was performed on the combination of all three types of features using RF. Subsequently, the RGB images were classified into three land-use types, including maize, bare land, and shadow. Bayes, KNN, SVM, DT, and RFs models were finally selected for the supervised classification with error matrix. The results showed that the optimal classification accuracy was obtained using neither a single feature nor all three types of features in total. More importantly, a combination of features was usually achieved higher accuracy than that of a single feature. Specifically, the best choice was the combination of HSI color and RGB image texture features using the RF, particularly with the total highest accuracy of 0.862 and a Kappa coefficient of 0.793. Additionally, the dimension reduction of features using RF models was neither significantly improved nor reduced classification accuracy (except for the SVM). However, the factors retained from the feature dimension reduction were easily explained suitable for the actual background and meaning. Furthermore, both classification efficiency and stability were improved greatly during this time. The finding can provide a specific solution for the high-precision extraction of crop planting information using UAV RGB images [ABSTRACT FROM AUTHOR]

With the continuous development of hyperspectral remote sensing technology, it has been widely used in vegetation mapping. However, sparse vegetation canopy, soil background, and spectral similarity between different types of vegetation are still the main challenges for vegetation types mapping in arid areas. As a result, it is difficult to achieve reliable classification accuracy by using spectral or texture features separately. Generalized Normal Distribution Optimization (GNDO) is a new feature optimization algorithm, with advantages in quality and stability of feature extraction results, comparing to traditional optimization algorithms. But it has not yet been applied to select bands of hyperspectral data. In order to validate the feasibility of combining ZY-1 02D spectral and texture features to classify vegetation types in arid areas, to verify the effectiveness of the GNDO method for bands selection of hyperspectral data, and to explore the effects of feature selection methods and training sample numbers on the classification accuracy of vegetation mapping, different Wrapper Optimization methods, such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Grey Wolf Optimization (GWO), and GNDO, were applied to select spectral features for vegetation mapping, taking the area around Zongjia Town, Dulan County, Qinghai Province, China as the research area, then the band selection results of these methods were analyzed. Train sample set containing 30, 50, 100, 150, and 200 pixels per class were used to select bands and to train the classifier. Different methods (ALL (without bands selection), GA, PSO, GWO, GNDO) and different sizes of the trained sample sets (30, 50, 100, 150, and 200 pixels per class) were used to obtain 25 spectral feature-based classification data sets. Simultaneously, 8 texture features (mean, variance, homogeneity, contrast, dissimilarity, entropy, second moment) were extracted using the Gray-level co-occurrence matrix (GLCM) method, and selected on basis of distinguishability for vegetation types. Texture features (TEX) were combined with spectral feature-based classification data sets. The random forest classification method was applied to classify vegetation types for the classification data sets, and the classification accuracy of classification data sets was evaluated and compared. The results show that 1) the blue region (400-450 nm), the red edge region (700-750 nm), and the red region (600-650 nm) are the most sensitive to distinguish the vegetation types in the study area; 2) the GNDO200 achieved the highest overall classification accuracy (80.44%) among the spectral feature-based classification data sets, which was better than the classification accuracy (78.86%) using all bands (ALL200); 3) with the increase of training samples, the overall classification accuracy of each classification data set showed an increasing trend, the classification accuracy of different feature selection methods showed different reliance on the number of training samples; 4) image texture features significantly improved the classification accuracy, and the GWO200+TEX dataset had the highest overall classification accuracy (82.86%). This study could verify the potential of the ZY1-02D, the new hyperspectral satellite data, for the classification of vegetation types in arid areas, and provide an idea for the selection of spectral and texture features in hyperspectral vegetation mapping. [ABSTRACT FROM AUTHOR]

A recirculating aquaculture system (RAS) has widely been acknowledged as eco-friendly land-based aquaculture since only a limited amount of water is used in recirculation. There is also much greater potential in modern intensive aquaculture, compared with conventional fish production. A key feature of RAS is the high precise controlling of the water environment in the rearing tank and the feeding of rearing fish. Water environmental parameters (such as water temperature, and oxygen levels) can provide stable and optimal conditions for less stress and better growth during fish living. In addition, fish feeding plays a vital role in RAS, because starvation may impact the organizational structures of fish, whereas, overfeeding may threaten the survival of animals, as well as the water quality. Therefore, it is essential to investigate the strategy of fish feeding in RAS. Currently, the feeding events are mainly implemented by the feeding machine with predetermined and fixed feeding times and quantities, according to the experience rather than the real fish appetite. Thus, it is necessary to evaluate the fish behavior pattern and appetite in the feeding strategy for a higher performance of RAS. In this study, an innovative evaluation of fish feeding behavior was proposed using Fourier spectrum feature extraction and support vector machines (SVM) classification. The degree of surface ripple dispersion was considered to predict the fish feeding behavior, rather than the only movement of fish bodies in conventional cases. Firstly, a high-speed video camera was used to capture the process before and after feeding. The video images were then converted from RGB to HSV space, where the background including the water ripple was subtracted, and the contrast was enhanced via the thresholds of saturation and brightness. Secondly, fast Fourier transform (FFT) was adopted to transform the image information from the spatial to a frequency domain. The fluctuation level of ripples in the image was regarded as an indicator of fish appetite using the amplitudes in the frequency spectrum. Specifically, the frequency spectrum was characterized by a high frequency, short period, and a large sum of amplitudes, when the fishes presented high appetite, and vice versa. Thirdly, the database of feature vectors was generated using the self-developed loop filters with various diameters. Finally, the obtained feature vector was trained to SVM. The feature vector of the SVM model was then optimized, where the optimal kernel width and penalty coefficient were selected to efficiently judge the feeding desire of fish during the cross validation. Moreover, a comparison was conducted to verify the efficiency, among the proposed, optical flow, image texture, shape, and texture features. As such, the original interference condition was transformed into the judgment, indicating the better universality of RAS against the high fish density, unstable water surface ripple, and dark light in the system. A field test also demonstrated that a high accuracy of 99.24% was achieved in judging the feeding behavior of farmed fish in RAS. In practice, it is strongly recommended to take videos of farmed fish as the training data for the classifier of fish feeding behavior. Consequently, the improved RAS can also be incorporated into the feeding machine to supervise the real-time smart and precise feeding events in modern aquaculture. [ABSTRACT FROM AUTHOR]

Here, a computer vision on-line detection system was applied to Shiitake mushrooms (Lentinus edodes) during the hot air drying with the process control of Relative Humidity (RH). Four RH controls were used, including continuous fanning, constant target RH throughout the whole process, reducing the RH at different stages, and rapidly reducing the RH at the later stage. An image processing using region segmentation was selected to on-line measure the Ratio of Wrinkled Surface Area (RWSA), in order to separate the wrinkled areas from the non-wrinkled areas on the surface of shiitake mushrooms. The roundness in the image was used to represent the shape of the edge of shiitake mushrooms. The Cell Aspect Ratio (CAR) was used to quantify the degree of deformation in a cell of shiitake mushrooms, which was calculated by the pixel length of cells along the radial and vertical direction in the microscopic image. The appearance features (contrast and energy) of shiitake mushrooms were obtained by the Gray Level Co-occurrence Matrix (GLCM). A measurement was made for the changes of shrinkage, roundness, RWSA, and visible texture (contrast and energy) characteristics during the drying process of shiitake mushrooms, and thereby to establish the relationship among appearance qualities, Rehydration Ratio (RR), and microstructure of shiitake mushrooms. The results showed that the RH had significant effects (P < 0.05) on the appearance qualities (shrinkage, roundness, RWSA, contrast, and energy), the microstructure, and rehydration ratio of shiitake mushrooms. Compared with the continuous fanning, higher RH maintained in the drying room at the initial stage of drying, corresponding to higher internal temperature of shiitake mushrooms, while longer drying time. The higher the initial RH or the longer the RH holding time was, the lower the drying rate and surface hardening degree were. Therefore, the shiitake mushrooms were more sensitive to shrinkage and deformation. The roundness changes of shiitake mushrooms corresponded to the change of shrinkages, where the higher the shiitake mushrooms shrinkage was, the lower the roundness was, indicating the more irregular shape of shiitake mushrooms. The slow rate of surface hardening led to more wrinkles on the surface of shiitake mushrooms, and lower color contrast distribution in the neighborhood pixels on the images, indicating that the color of non-wrinkled area was close to that of wrinkled area, and the whole surface color of shiitake mushroom was darker. Meanwhile, the cell aspect ratio of shiitake mushrooms cells was larger, indicating that the deformation of cells was more serious. At the same time, the small cell pore resulted in very low rehydration ratio of shiitake mushrooms. Compared with the whole process groups, the drying time, and shrinkage were significantly (P < 0.05) reduced, the roundness improved significantly (P < 0.05), the RWSAs reduced (P > 0.05), and the surface color of shiitake mushrooms lighter (P < 0.05) in the optimized groups by turning on the dehumidification fan, when the moisture evaporation was not enough to maintain the set RH range in the drying room. The reason can be the increased differential partial pressure between the air and the surface of shiitake mushrooms, resulting in an accelerated drying rate, while rapidly hardened and shaped surface of shiitake mushrooms. However, the deformation degree of shiitake mushroom cells was not significantly (P > 0.05) reduced, and there was no increase in the erehydration ratio. A positive correlation was observed between the time, when the RWSA reached the maximum and the rehydration ratio of shiitake mushrooms (r = 0.88). High relative humidity had a negative effect on the flavor components of shiitake mushrooms. The best drying was continuous fanning, according to the drying time, appearance quality, sensory quality, and composition of dried shiitake mushrooms. [ABSTRACT FROM AUTHOR]

The wide use of plastic greenhouses in the word has brought economic benefits, but also caused many environmental problems. Accurate and timely acquisition of spatial distribution information of plastic greenhouse is of great significance to agricultural production and soil management. The use of plastic greenhouse has changed the structure of soil, thus changing the surface spectral characteristics of the soil. Because of the particularity of the plastic film material, the reflection spectrum has strong directionality and uncertainty, so it is difficult to identify plastic greenhouse accurately only depending on the reflectance spectrum characteristics of ground objects. In this paper, Kunming City, Yunnan Province is taken as the research area, using GF-2 image as a single data source, the multi-scale segmentation method is used to obtain the image object efficiently and accurately. According to the detailed spatial information of plastic greenhouse, the applicability of three image object-based texture extraction algorithms, namely GLCM(gray-level co-occurrence matrix), LBP(local binary pattern) and PSI(pixel shape index) for plastic greenhouse identification is analysed and tested. In addition, different texture features are combined with spectral features and NDVI index to form different classification schemes, to explore which one would be the best combination of texture features for identification of plastic greenhouses. In order to explore the robustness of the method, different texture feature combination schemes are applied in the study areas. The results of SVM (support vector machine) classifier are evaluated by confusion matrix. The results show that the overall combination schemes of the two research areas with different landscape patterns have the same trend. The combination of spectral features and NDVI index can accurately identify the scattered waters in each of the two study areas. For plastic greenhouses and impervious surfaces with similar reflectance spectra, adding texture features can make up for limitation of spectral characteristics and improve the overall accuracy. The phenomena of homologous or homologous spectra in spectral features of high spatial resolution image can effectively improve the discrimination between plastic greenhouse and impervious surface. Texture features can significantly improve the identification accuracy of plastic greenhouse following the object-based image classification frame. In the classification scheme of plastic greenhouses with different spatial distribution structures, the LBP (local binary pattern) texture algorithm has the best recognition accuracy, the overall accuracy of study area A is 96.85%, Kappa coefficient is 0.95, and that of study area B is 95.58% and 0.94. Landscape fragment analysis (landscape fragmentation index area mean index, aggregation indices) of the two different study areas showed that the plastic greenhouses in study area A are more fragmented than study area B (area mean indices are 3.39 hm² and 1.37 hm², aggregation indices are 80.64 and 72.98 for plastic greenhouses in study area A and B, respectively). The results of fragmentation are consistent with those of classification, and the accuracy of landscape classification with more space fragments is lower (the highest overall classification accuracy for study area A and B are 98.13% and 96.13%, respectively, the PA(producer accuracy) and UA(user accuracy) are 96.47% and 97.93% for study area A, and 90.67%, 99.68% for study area B). The results show that texture features based on image objects can improve the recognition accuracy of small-scale plastic greenhouse. This is of great significance to the accurate mapping of the distribution of plastic greenhouses. [ABSTRACT FROM AUTHOR]

The wide use of plastic greenhouses in the word has brought economic benefits, but also caused many environmental problems. Accurate and timely acquisition of spatial distribution information of plastic greenhouse is of great significance to agricultural production and soil management. The use of plastic greenhouse has changed the structure of soil, thus changing the surface spectral characteristics of the soil. Because of the particularity of the plastic film material, the reflection spectrum has strong directionality and uncertainty, so it is difficult to identify plastic greenhouse accurately only depending on the reflectance spectrum characteristics of ground objects. In this paper, Kunming City, Yunnan Province is taken as the research area, using GF-2 image as a single data source, the multi-scale segmentation method is used to obtain the image object efficiently and accurately. According to the detailed spatial information of plastic greenhouse, the applicability of three image object-based texture extraction algorithms, namely GLCM(gray-level co-occurrence matrix), LBP(local binary pattern) and PSI(pixel shape index) for plastic greenhouse identification is analysed and tested. In addition, different texture features are combined with spectral features and NDVI index to form different classification schemes, to explore which one would be the best combination of texture features for identification of plastic greenhouses. In order to explore the robustness of the method, different texture feature combination schemes are applied in the study areas. The results of SVM (support vector machine) classifier are evaluated by confusion matrix. The results show that the overall combination schemes of the two research areas with different landscape patterns have the same trend. The combination of spectral features and NDVI index can accurately identify the scattered waters in each of the two study areas. For plastic greenhouses and impervious surfaces with similar reflectance spectra, adding texture features can make up for limitation of spectral characteristics and improve the overall accuracy. The phenomena of homologous or homologous spectra in spectral features of high spatial resolution image can effectively improve the discrimination between plastic greenhouse and impervious surface. Texture features can significantly improve the identification accuracy of plastic greenhouse following the object-based image classification frame. In the classification scheme of plastic greenhouses with different spatial distribution structures, the LBP (local binary pattern) texture algorithm has the best recognition accuracy, the overall accuracy of study area A is 96.85%, Kappa coefficient is 0.95, and that of study area B is 95.58% and 0.94. Landscape fragment analysis (landscape fragmentation index area mean index, aggregation indices) of the two different study areas showed that the plastic greenhouses in study area A are more fragmented than study area B (area mean indices are 3.39 hm2 and 1.37 hm2, aggregation indices are 80.64 and 72.98 for plastic greenhouses in study area A and B, respectively). The results of fragmentation are consistent with those of classification, and the accuracy of landscape classification with more space fragments is lower (the highest overall classification accuracy for study area A and B are 98.13% and 96.13%, respectively, the PA(producer accuracy) and UA(user accuracy) are 96.47% and 97.93% for study area A, and 90.67%, 99.68% for study area B). The results show that texture features based on image objects can improve the recognition accuracy of small-scale plastic greenhouse. This is of great significance to the accurate mapping of the distribution of plastic greenhouses. [ABSTRACT FROM AUTHOR]

Chlorophyll is an important pigment for crop light energy utilization, which directly affects the process of energy material conversion and transmission. The change of chlorophyll content directly reflects the ability of photosynthesis and the nutritional status of crop growth. When the traditional unmanned aerial vehicle (UAV) remote sensing was used for crop nutrition monitoring, most of them started from the spectral vegetation indices, ignoring the characteristics of the image itself. In this study, we estimated potato leaf chlorophyll content from a comprehensive index formed by the fusion of multi-spectral vegetation indices, texture features and comprehensive indicators of data fusion. The effect of comprehensive index model on estimating potato leaf chlorophyll content was explored. First, we used the UAV multi-spectral images during the whole potato growth period in 2018 in Xiaotangshan, Changping, and Beijing. The multi-spectral vegetation index, texture characteristics and other variables were first extracted from UVA images, then their correlation relationships with leaf chlorophyll content were analyzed. The optimal image variables were screened out, and the whole subset analysis was based on adjusted determination coefficient and 10-fold cross-validation was used to estimate the leaf chlorophyll content of potato. Finally, the vegetation index and texture features were reconstructed by principal component fusion to establish a new comprehensive index for chlorophyll content estimation. It was found that the leaf chlorophyll content estimation model based on comprehensive index was better than that based on multi-spectral vegetation indices and texture features. The main reason was that the comprehensive index contained both spectral information and image texture information. Multispectral information and model accuracy had also been significantly improved. In the bud period, compared with the vegetation indices based model and the texture feature based model, the determination coefficient (R2) of the comprehensive index model increased 0.104 and 0.136, while the normalized root mean squared error (NRMSE) reduced 1.3% and 1.6%. During the tuber formation period, the determination coefficient of comprehensive index model was increased 0.04.and 0.101, while the NRMSE was decreased 0.5% and 1.2%, compared with the other 2 models. In the tuber growth period, the determination coefficient of comprehensive index model increased 0.075 and 0.111, and the NRMSE decreased 0.9% and 1.3% compared with the vegetation index model and the texture feature model. During the starch accumulation period, the R2 of comprehensive index model increased 0.017 and 0.046, and the NRMSE decreased 0.2% and 0.6%, compared with vegetation index model and texture feature model. In the maturity period, the determination coefficient of comprehensive index model increased 0.092 and 0.057, and the NRMSE decreased 2.4% and 1.5%.Therefore, the effect of comprehensive index estimation model was the best followed by multi-spectral vegetation indices model and the texture feature model was the worst. The starch accumulation period was the best growth period for estimating chlorophyll content by multispectral vegetation index and texture characteristics, while bud period was the best growth period for estimating chlorophyll content by comprehensive index. Estimating potato chlorophyll content based on multi-spectral image from UAV platform can provide a feasible method for potato growth nutrition monitoring. It realizes low-cost, fast and high-throughput monitoring of potato growth and nutrition information, as well as provides guarantee for fine management of farmland irrigation, variable fertilization and so on. [ABSTRACT FROM AUTHOR]

In the machine vision-based intelligent system for recognizing female or male silkworm pupa, much spatially-varying blur appears in silkworm pupa images and it could give rise to the loss of images textures and structures to a great extent due to the irregular ellipsoid shape of silkworm pupa and the limited depth of field of optical imaging system. This brings a challenge for an intelligent system to identify silkworm pupa’s gender. Shen’s method is supposed to be one of the state-of-the-art methods, but the PSF(point spread function) is estimated on a per pixel basis and parameter enumeration is required to meet the optimization criterion, which leads to a prohibitively large computation efforts. To solve this problem, we presented an effective method that the complicated restoration of spatially-varying blur silkworm pupa images was decomposed into the simple restoration of multiple images which have the same level blur and were part of original image. In this work, according to the variation of Tang’s method, the blur standard deviation at every pixel was estimated to construct a full defocus map of silkworm pupa image. The approximate blur standard deviations in defocus map were automatically sorted via the fuzzy C-means algorithm, and the original blur silkworm pupa image, based on this classification, was naturally segmented into several sub images possessing similar level blur. Then, each sub image was deblurred by using Lucy-Richardson (LR), and was merged to form a full silkworm pupa image. Eventually, Bilateral Filtering was used to eliminate the errors arising in the merging stage, and the high-quality silkworm pupa image was then obtained. To test this method, experiments (including both female and male silkworm pupa images) were conducted on the platform configured with CPU i5-2430 M, 2.4 GHz, memory 2 G, 32 bit operation system, matlab 2012 and VC++6.0. We set iteration steps of LR de-convolution as eight in real data experiments. Total variation Mean (TVM) was used to estimate the quality of the restored results. The experimental results showed that the performance of the proposed algorithm was better than Shen’s method. The method successfully removed spatially-varying blur and enhanced the image quality, which was verified in both qualitative (or visual) and quantitative ways. It can be seen that in real data experiments, our method effectively improved silkworm pupa images from which the spatially-varying blur was eliminated to a great extent, more image texture details were increased and sharpness contrast was much better. Meanwhile, in term of quantitation estimation, the TVM values of our method’ results were bigger than Shen’s results, which was further proof of our method’s good performance. It was noted that our method can also be conveniently extended to improve the quality of other vegetable images suffering from the spatially-varying blur, such as marrow and pumpkin, as shown in the experiments. After silkworm pupa image restoration, we achieved high accuracy of 92.3% in identifying male and female silkworm pupa. The proposed method can have a wide application of machine vision technologies. [ABSTRACT FROM AUTHOR]

Recently, it's still difficult to entirely replace the artificial visual interpretation for the computer automatic classification, which is used to extract land cover types' information from the remote sensing imagery, because the automatic method needs more efforts to improve the precision of the classification results. Furthermore, this problem has become the key joint of the automatic classification extraction. How to extract land cover types' information in western area of Jilin, is one of the important problems, and the confused land cover types needs to be distinguished. The aim of this study is to deepen the application of remote sensing classification method that is used to extract land cover information automatically and quickly from the satellite imagery. The western area of Jilin is selected as the main research area. A new total solution to extract land cover information, based on the spatial variation theory, has been designed for the convenient automatic classification with the remote sensing technology. The remote sensing classification scheme is carried out by coding the R language algorithm and operating the remote sensing software ERDAS platform. The land cover types in Zhenlai County in the western area of Jilin, have been extracted and monitored through the combined utilization of 4 indices, including semivariance value of normalized difference vegetation index (NDVI) dataset, local variance of image texture, modified soil-adjusted vegetation index and normalized difference water index, which have significant meaning for the land cover types in the transition zone between cropping area and nomadic area. These variances have definite physical meaning (including vegetation, water, and soil drought conditions), so that the phenological information was used to build a multi-dimensional feature space classification data set. The results indicated that: 1) A total of 11 land cover types are extracted, using the multi-temporal remote sensing information to build a multidimensional classification characteristics data set based on the Landsat 8 data. The overall classification accuracy of the algorithm is 95.50%; the Kappa coefficient of classification is 0.9504. The automatic extracting approach implemented obtains a comparatively ideal classification result; 2) The introduction of 3 characteristic variables of the classification in the scheme significantly improves the separability of the confused land cover types. Considering the vegetation classification, the vegetation growth information has practical life-activity significance, and is a real-time dynamic method for the vegetation change monitoring; 3) Improving the land cover classification accuracy is not to introduce more characteristic parameters of the classification, but to effectively combine multiple appropriate classification variables. The new method can broaden the application vision and the scope of the ecological remote sensing investigation of surface vegetation. Moreover, the introduction of new variables not only makes the macro monitoring more convenient, but also improves the accuracy of classification of remote sensing interpretation. It's noted that the extracted classification has obvious regional feature, and the regional feature is consistent with the farming cultivation characteristics on the Northeast Plain. In a word, the results can provide a credible approach and valuable example for extracting and monitoring land cover type in farming-pastoral transitional zone. It is feasible to use the spatial variation theory to extract and monitor land cover type by combining the several evaluation indices. [ABSTRACT FROM AUTHOR]

In order to conquer the difficulty of simultaneously recognizing the multiple defects of potatoes samples placed randomly, this paper proposed a non-destructive detection method which combined manifold learning dimension reduction algorithm based on hyperspectral information fusion and extreme learning machine (ELM) to simultaneously distinguish the multiple defects of potatoes. In this paper, 367 potatoes were picked which were made up of 111 sprouting potatoes, 90 green rind potatoes, 46 blackheart potatoes and 120 normal potatoes. The hyperspectral image acquisition system contained imaging spectroradiometer (SPECIM, V10E, Finland), data acquisition box, lighting system, electric moving stage and objective table. The reflection hyperspectral information of all those potatoes was acquired by using the hyperspectral image acquisition system, whose spectral wavelength ranged from 390 to 1040 nm. The hyperspectral information included the spectral information from 520 wave bands and the imaging information from 520 gray images. After the correction to hyperspectral data with the standard black and white board, hyperspectral data in the range of 450-990 nm, which had high signal-to-noise ratio (SNR), were selected as original spectrum for subsequent processing. To deal with the spectral information, the average spectrum was abstracted from the region of interests (ROI) on every potato sample by using the environment for visualizing images (ENVI). Comparing several data preprocessing methods, detrend was determined as the optimal spectral preprocessing method. Diffusion maps (DM), locally linear embedding (LLE) and hessian locally linear embedding (HLLE) were respectively utilized for the purpose of cutting down the dimension of spectrum data after the spectral preprocessing named detrend. To deal with the hyperspectral imaging information, every pseudo-color image of potatoes was morphologically processed before extracting 84 image texture characteristics based on gray level co-occurrence matrix (GLCM). By using successive projections algorithm (SPA), 10 texture features were properly selected, which consisted of homogeneity of 45° GLCM in R gray image, energy of 90° GLCM in R gray image, correlation of 90° GLCM in R gray image, contrast of 135° GLCM in R gray image, correlation of 0° GLCM in G gray image, contrast of 0° GLCM in G gray image, correlation of 90° GLCM in G gray image, homogeneity of 45° GLCM in B gray image, correlation of 90° GLCM in B gray image and contrast of 135° GLCM in B gray image. After the fusion of the spectral characteristics and image features, the 367 potato samples were divided into calibration set and test set. The calibration set had 250 potato samples including 74 sprouting potatoes, 59 green rind potatoes, 28 blackheart potatoes and 89 normal potatoes. The whole number of samples in test set was 117 which contained 37 sprouting potatoes, 31 green rind potatoes, 18 blackheart potatoes and 31 normal potatoes. Six models were respectively established based on support vector machine (SVM) and ELM, which were DM-SVM, LLE-SVM, HLLE-SVM, DM-ELM, LLE-ELM and HLLE-ELM. Comparing and analyzing the results of the 6 models and the spending time, we discovered that the DM was the best manifold learning dimension reduction algorithm to deduce the spectral information. We alsofound that the DM-ELM model was the best model because of the better recognition rate for recognizing the multiple defects of potatoes and spending less time. For the DM-ELM model, the single recognition rate of sprouting potatoes, green rind potatoes, blackheart potatoes and normal potatoes respectively reached 97.30%, 93.55%, 94.44% and 100%, and the mixed recognition rate reached 96.58%. It only cost 0.11 s to build the DM-ELM model. The results indicate that combining hyperspectral information fusion with manifold learning dimension reduction algorithm can simultaneously distinguish the multiple defects of potatoes. [ABSTRACT FROM AUTHOR]

Both artificial visual interpretation and computer automatic classification were the mainly remote sensing methods used to extract farmland information. At present, it is still difficult to entirely replace the artificial visual interpretation for the computer automatic classification to extract farmlands' type information of the remote sensing image, because the automatic method needs more efforts to improve the precision of the classification results, so the problem became the key link of the automatic classification extraction. How to extract farmlands' type information in the western part of Jilin is one of the major problems which the paper attempts to solve to distinguish paddy field from dry land. A new solution to extract the farmland information has been designed for the remote sensing automatic classification, based on the spatial variation theory. The classification scheme was carried out by operating in an R language platform and the remote sensing software ERDAS platform. The farmland type of Zhenlai in the western of Jilin was extracted and monitored by making use of four indexes, the range of NDVI series, the local variance of image texture, the modified soil adjusted vegetation index, and the normalized difference water index, which have significant meaning for the farmland cover type in the transition zone between the cropping area and the nomadic area. These variances with clear physical meaning information (including the vegetation, water, soil drought conditions) and phonological information were used to build a multi-dimensional feature space classification data set. The results indicated that: 1) the dry field of which an area of 1065.337 km2 was cultivated in the study area was the largest farmlands' type, and also was one of the most important ecological landscape types. It's the spatial distribution characteristic of the study area that is a relatively dispersed dry field, and a relatively concentrated paddy field; 2) based on the multidimensional space data set, the algorithm of a support vector machine (SVM) was chosen to automatically extract the farmland types' data of the paddy field and the dry land. The overall classification accuracy of the algorithm was 94%, the Kappa coefficient of the classification was 0.87, and the extracted accuracy of the paddy field was 98.3%, while the extracted accuracy of the dry land was 98%. The existing automatic extraction approach was implemented to obtain a comparatively ideal classification result; 3) through the farmland's regional analysis, a depression that has more lowland and easy seeper is suitable for the reclamation of paddy field. It's noted that the extracted classification has an obvious regional farmland type, and the regional features were consistent with the farming cultivation characteristics in the northeast plains; 4) both rice and corn were typical unimodal type growth crops, and the similar growth peak, but the result of the range (rice, 3.8±0.4; upland crops, 3.4±0.3;) noted that: the phenology information of different vegetation types has its own characteristics; this characteristic of the NDVI seasonal variation curve is real. [ABSTRACT FROM AUTHOR]

Due to high protein, low fat, high vitamin and mineral content, beef is regarded as an important meat item which is consumed by human in regular basis. In addition, with the improvement of people's living standard, different quality parameters such as beef tenderness, water retention and other edible quality indicators have become important factors to consumers' satisfaction. Sensory evaluation and shear force measurement are primarily two methods for measurement of beef tenderness, which are now considered as traditional methods. However, with improvement in technology, optical methods are playing important role for rapid, real-time, non-destructive and accurate detection of meat quality parameters. Near-infrared method and hyperspectral imaging method are important methods for non-destructive measurement of beef tenderness. This study focused on the application of hyperspectral image analysis for non-destructive detection of beef tenderness at different points of the sample. Considering the measurement complexity due to complex beef structure as well as the disadvantage of near infrared spectrum for single point detection, this study made use of hyperspectral imaging system to obtain three-dimensional information from each sample. Fifty-six beef samples were obtained from the slaughter house Yuxiangyuan in Beijing and were cut into 3 cm thick slices, vacuum packaged and then stored in 4°C environment before hyperspectral image acquisition. Reflectance spectrum was obtained for each sample from acquired images. Stepwise regression and genetic algorithm (GA) were used to identify the feature wavelengths, thus reducing the high dimensionality of the hyperspectral data. Eight feature wavelengths were selected, which were used to develop prediction model for measuring the shear force value at each point in the sample. By the principal component analysis (PCA), every sample could acquire 5 PCA images. Gray level co-occurrence matrix (GLCM) was applied to extract 8 main texture parameters from the characteristic band image and the 3 PCA images. Eight wavelengths were extracted from these 8 texture characteristic parameters respectively, with a total of 64 texture variables. Also, from five PCA images 40 texture variables were extracted. These parameters were used to establish a model based on linear discrimination analysis (LDA) and support vector machine (SVM) for beef tenderness. Because the number of principal components has a greater impact on the accuracy of discriminant model, before the training of SVM model, it first needed to select the appropriate number of principal components. Finally, three PCA images which included 95.37% information of every sample were chosen to extract texture parameters. With texture parameters of band images, the prediction set's accuracy of the discriminant model was 88.89%. There had some shortcomings that on one hand, the band images had great influence on the result, and on the other hand, some information may be missed with selected band images, so the texture features of the image of sample main component were used to establish classification model. After analysis and comparison, the accuracy of main component model was 94.44%, which was higher than that of band image model. In addition, the LDA method was more applied to establish classification model than SVM method. The results show that the image texture features can determine the distribution of beef tenderness. The results can provide the theoretical reference for the research on rapid non-destructive detection technique of beef tenderness. [ABSTRACT FROM AUTHOR]

An intelligent fruit and vegetable recognition system utilizing image recognition can accurately and rapidly indentify different kinds of fruits and vegetables, which can improve supermarket and market sales efficiency. The feature extracting method is a very important issue in an intelligent fruit and vegetable recognition system. However, traditional fruit and vegetable recognition algorithms either ignore the texture feature of fruits and vegetables, or used texture features that couldn't better represent the texture of fruit and vegetable images. In order to represent the texture feature of fruit and vegetable images better and improve the intelligent fruit and vegetable recognition system recognition rate, we proposed a novel texture feature extraction algorithms called color completed local binary pattern (CCLBP) in this paper. By extracting different kinds of color channels completed by a local binary pattern (CLBP) texture feature, the CCLBP constructed a new texture feature extraction algorithm. The Fruit and vegetable recognition system model uses CCLBP to extract an image texture feature, and uses a HSV color histogram and Border/interior pixel classification (BIC) color histogram to extract image color features. Then it uses a matching score fusion algorithm to fuse color and texture features, and finally, a nearest neighbor (NN) classifier is used to realize fruit and vegetable recognition. To verify the effectiveness of the algorithms, two different fruit and vegetable databases, called an interior database and an outdoor database, were constructed in this paper. The interior database acquired in a laboratory contains 13 kinds of fruits and vegetables, which is used to verify algorithms recognition performance under different kinds of illumination. The outdoor database acquired in the market contains 47 kinds of fruits and vegetables, which is used to verify algorithms recognition performance under a different number of training sets. A Fruit and vegetable recognition experiment under different kinds of illumination showed that, only by the texture feature indentifying the kinds of fruits and vegetables, the recognition rate of the CCLBP was 5% higher than the traditional fruit and vegetable texture features (such as Unser, TestA), which means that the CCLBP is more suitable for fruit and vegetable recognition; besides, compared with other texture algorithms, the CCLBP fused with HSV color histogram and BIC color histogram can achieve a 73.93% highest mean recognition rate, which takes about 1.1 seconds indentifying an image. A fruit and vegetable recognition experiment under a different number of training sets had similar results as the experiment under different kinds of illumination. The recognition rate of the CCLBP was still higher than the traditional fruits and vegetables texture features. What's more, the CCLBP fused with a HSV color histogram and a BIC color histogram can achieve 94.26%, the highest recognition rate. The experiments under different kinds of illumination and under different number of fruits and vegetables confirm the feasibility of our algorithm. Our algorithm can be used in intelligent fruit and vegetable recognition system, which improves the system accuracy rate. [ABSTRACT FROM AUTHOR]

Crop disease is one of the main disasters for Chinese agriculture and it seriously affects the yield and quality of crops, and causes economic losses to farmers. Early detection and prevention of crop diseases is critical to control the diseases, improve crop yields, reduce the economic losses and control pesticide pollution. Therefore, the research of recognition methods for crop diseases is necessary. In this study, a disease recognition method of cucumber disease, based on leaf image and environmental information, is proposed. In this method, the cucumber disease features, including environmental classifying features and disease leaf classifying features, were extracted by image processing and statistical analysis methods. The classifying features were then selected by SAS discriminant analysis, and the cucumber diseases were identified by using the rule of maximum membership degree. The cucumber leaves and their environmental information of Downy mildew, Brown Speck, and Anthracnose were collected for disease recognition. The diseased cucumber leaf images were processed by using a series of image pre-processing methods, such as image transforming, smoothing, and segmentation. White was chosen as the background of the diseased leaf images, a median filter was applied to effectively remove the disturbance of noise, and the color image segmentation method, based on statistical pattern recognition, was applied to separate the disease spot images from the diseased leaf images. The five features of environmental information were extracted, and the 35 statistical eigen vectors of color, shape, and texture of the diseased leaf images could be extracted by statistical analysis. Then, 40 disease union classifying features were obtained. Ten strong classifying features were then selected by the SAS discriminant analysis method. The feature vectors of the clustering center were then computed. Finally, three kinds of cucumber diseases were recognized by the maximum membership degree. The disease recognition method proposed in this study differs essentially from the traditional ones. Traditional methods only take into account features extracted from diseased leaf images, which makes the recognition rate of traditional methods low because the diseased leaf image is quite complex. However, the proposed disease recognition method contains not only the leaf image information, but also the environmental information, which improves the robustness and recognition rate of the proposed method. The recognition results of three kinds of cucumber diseases by the proposed method were more than 93 percent. The experiment results show that cucumber diseased leaf images obtained under different conditions were effectively recognized by the integrated application of image processing technology, analysis of image texture, color and figure characteristics, and analysis of the disease environmental information, etc. It has provided a technical basis and support for the automatic recognition of crop diseases with diseased leaf images and environmental information obtained in the field. The analysis and experimental results in this paper demonstrated that the crop disease recognition method is feasible by computer vision, statistics, and comprehensive utilization technology of spot color, texture, shape information, and crop environment information. As there are many factors affecting crop diseases, various diseases in different periods will show different symptoms. Therefore, how to use the computer vision technology and the disease environmental information to build a powerful and practical crop disease recognition method still needs further study. [ABSTRACT FROM AUTHOR]

In order to solve the problem of muskmelon fruit image segmentation under a complex background, an algorithm of image segmentation based on fusing color feature and texture feature was proposed in this paper. First, the collected muskmelon fruit images were transformed from RGB color space to CIELAB color space and HSV space respectively. According to the color characteristics of muskmelon fruit, the collected images were binarized using the threshold of angle model that was set up in using a*b* components in CIELAB color space. To reduce the influence of the uneven illumination distribution of segmentation, the H S components segmentation threshold was selected to binarize the collected images. Converging the results of the angle model segmentation and the HS weighted threshold segmentation, the results were obtained based on color feature segmentation. Then, the image texture features were extracted and the binarization images were obtained by using texture feature threshold. The segmentation results were achieved by fusing the texture features and the color feature segmentation result. Finally, taking the fruit color feature segmentation area as the qualification, the final segmentation results were obtained by binding growth based on the segmentation area that were obtained by fusing color features and texture features. In order to evaluate the effect of proposed algorithm, the collected images were segmented using the super green threshold algorithm and the NDI algorithm and the results were gained. The average detection rate of three algorithms were 83.24%, 43.12% and 99.09%, respectively. Comparing the results of the detection rate and false detection rate, the experimental results of the proposed algorithm were superior to super green feature segmentation and normalized difference index (NDI) segmentation algorithm. [ABSTRACT FROM AUTHOR]