فهرست مطالب abdolabbas jafari

The improper use of chemical fertilizers in crop production can result in unsafe food sources for consumers. This research focuses on estimating the accumulation of nitrate in tomatoes by analyzing images of tomato tissues. The experiments were conducted using a completely randomized design with four nitrogen levels: 400, 800, 1200, and 1600 kg.ha-1. Fifty samples were randomly selected from each treatment to create images for feature processing and develop a prediction model. The samples were sliced to a consistent thickness, and their images were prepared. The nitrate contents of the same samples were then measured in the laboratory. Color features, including R, G, and B color components, as well as non-color features such as white pixel area (WPA), total slice surface area (TSA), and the ratio of white pixel area to total slice surface area (WPA/TSA), were extracted from the images. The results showed that the nitrate content of the samples increased significantly (P<0.05) in response to the applied nitrogen fertilizer, with measurements of 1.6%, 2.7%, 2.8%, and 3.3%, respectively. Moreover, a strong correlation was found between the color components, WPA, TSA, WPA/TSA, and nitrate accumulation in the samples. Multiple regression and multilayer perceptron neural network (MLP) models were employed to predict the nitrate content. The best subset method was used to build an appropriate regression model. Various topologies and transform functions were applied to identify the best MLP model. The results indicated that an MLP model with a 3-15-1 topology and the lowest mean relative percentage error (MRPE) was the most accurate neural network model. The final regression and neural network models were validated using 60 intact samples. The neural network model achieved a MRPE of approximately 3.5%, demonstrating its precise estimation of nitrate contents compared to the regression model with an MRPE of around 5.2%.

In this research, the ability of the integrated non-destructive system including ultrasound, electrical resistance measurements and colorimetric assays to predict the textural properties (hardness, brittleness, viscosity, elasticity, guminess and chewiness) of trout fillets were evaluated. For this purpose, fillets were stored for 12 days at refrigerator temperature and then examined at interval times. At the same time physical, mechanical, chemical and sensory tests were performed on the fillets. The performance of neural network and support vector machine methods for predicting and modeling tissue properties were compared. In each model, physical properties were considered as inputs and textural properties as outputs and modeling was performed. The results showed that in the prediction of hardness, fragility and elasticity of the support vector machine and in the viscous, resin and chewability indices, the neural network method had more capability to model the texture properties, so that the root mean square error (RMSE) Hardness, fragility, viscosity, elasticity, gum and chewability indices were equal to 0.144, 0.025, 0.015, 0.015, 0.044 and 0.171 respectively and their correlation coefficients were equal 0.993, 0.985, 0.992, 0.961, 0.995 and 0.995. Therefore, the proposed non-destructive system in combination with artificial intelligence methods showed promises as a non-destructive and efficient tool for monitoring and quality control during trout fillet storage.

Separation and weighing of broken of rice kernels within a batch is a common way for measuring the amount of brokens. Separation is carried out by sieve or hand that is time-consuming method and required periodical sampling. In the present study, some image texture features of a rice batch for determining the percentage of brokens were applied. In order to this, twenty-one samples of 100 g were prepared with different breakage percentages from mixture of broken and whole rice kernels (containing zero to 100 percent broken). Ten images were then taken from each sample batch. After pre-processing of the images such as noise removing and creating binary images, the texture features of the images were extracted using MATLAB software. Twenty-three texture features were extracted using a grayscale co-occurrence matrix method. The principal components analysis (PCA) method was employed to reduce number of the features. By which, 23 calculated and measured texture features was reduced to three principal components (latent variables). Then, the relationship between percentage of broken rice in the sample and the principle components of 23 extracted texture features from the images by artificial neural network method was pinpointed. Four different topology of neural networks with different neurons in two hidden layers were employed. The results of neural networks outputs revealed that 3-6-8-1 topology could determine the percentage of broken in a batch of rice kernels with 0.958 coefficient of determination (R2) and total estimation error of 1.5 percent. Moreover, it was found that 10 texture features were related to first principal component, four features related to second component, four related to third component, and five fractures were simultaneously related first and second principal components. Eight features with correlation more than 0.9 with the first and third principal component had the most impact on estimation of percentage of broken. This method of measurement is fast, affordable and has proper precision with respect to the common (hand) method for online applications.

The germination percentage of different sugar beet cultivars and hybrids in the field is of special importance for the breeder as a superior feature of the improved cultivar. The aim of this study was to provide an image processing-based method for rapid and accurate counting of seedling number of various sugar beet cultivars in the field. The images were taken using a digital camera from a fixed height and transferred to the Matlab software environment in May 2017 at Zarghan Agricultural Research Station. Using image processing techniques and functions in the software, the algorithm for counting and determining the number of sugar beet seedling in a fixed length was coded and presented. The proposed algorithm was related to situations in which the leaves of sugar beet seedling were far from each other or only in contact with each other. The accuracy of the algorithm was 90.32% and its execution time was about 1.67 seconds. The actual number of seedlings and the number observed by the algorithm were statistically analyzed by paired t-student test and it was found no significant difference between them (P<0.05). Also, different sugar beet cultivars had no effect on the implementation of the algorithm and its accuracy. The results showed that instead of visually counting the number of seedlings, this procedure can be done with more than 90% accuracy for the above two conditions by imaging and using the algorithm.

A practical model including a parameter as a function of drying condition were introduced to predict the effect of far-infrared radiation (FIR) in a combined rough rice drying process. Tests were conducted in three replication for four levels of FIR (0, 0.1, 0.2 and 0.3 Wcm-2) and three levels of inlet air temperature (30, 40 and 500C). The results obtained from the simulation of the proposed model demonstrated that the models were capable satisfactorily express the characteristics of drying rough rice in combined dryer. Also due to the effects of combination of hot air and FIR drying technique on drying rates, cracking enhancement ratio (CER), required breaking energy for grain, specific energy consumption (SEC) and optimal method was presented. Combination of hot air at temperature of 40 0C and FIR at intensity of 1000 W.m-2 is the best way to have lowest SEC and highest grain quality point of view.

In the present study, classification of tomatoes by fuzzy logic and change in classification due to storage conditions were studied. In order to conduct the study, qualitative properties such as color, size and hardness of tomatoes were measured. Tomato samples were sorted by a deigned fuzzy logic system based on fuzzy rules, and compared with the experts. Sensitivity analysis of outputs of membership functions were tested by chi-square statistic. In cool room storage, sensitivity of membership functions started from six-day after storage with change from grade 1 to grade 2. For samples that kept at ambient temperature sensitivity started on third day after storage with the same change in sorted groups as previous tests. The results emphasized that any processing operation or marketing after initial sorting of tomatoes should spend time in a way that does not change the results of the initial classification.

Using the intelligent pruning machine can be reduces the labor required for pruning. It was attempted in this study to develop an algorithm that uses stereo vision techniques to identify which parts of the grapevine should be cut. Photos were taken from gardens of the Research center for Agriculture and Natural Resources of Fars province. At first plants were segmented from the background then the main trunk and one year old branches were identified and pruned based on the criteria for pruning. Then the main skeleton of grapevine was determined. Using this skeleton, the attaching points of the branches were obtained as well as on the trunk. Distance between the branches was maintained. Then the algorithm has been evaluated, the evaluation results of algorithm showed that the proposed algorithm had acceptable performance and pruning point of the grapevines were correctly detected. The accuracy of the developed algorithm was 96.8 percent.

In this research, the ability of a machine vision system was evaluated to monitor the oxidation period of olive oil. For this purpose, the olive oil samples studied during accelerated storage for 24 days in an oven and then image processing processes was carried out to extract parameters related to color spaces RGB, HSV and L*a*b*. The performance of artificial neural network and decision tree techniques compared to determine the olive oil oxidation rate. In each of the models color parameters were used as inputs and different stages in the olive oil oxidation period were considered as output. According to the results, the highest classification accuracy (94.44%) and the lowest RMSE (0.0696) are related to the decision tree technique. The proposed machine vision system combined with artificial intelligence techniques as non-destructive and efficient tool will offer for monitoring and quality control during oil storage.

Grading science and grading equipment for many kinds of sea products are growing rapidly in developed communities and variety of grading equipment can be found in most of the large fishery units. Computer vision has the potential to be used as a precise method for recognition and assessment of apparent characteristics. In this study، machine vision technology was used to sort fish based on species، size and weight. Tiger-toothed croaker and Silver pomfret fishes were selected for this study. In the first stage، each sample fish was weighted and put in the illumination chamber and images were captured. Matlab environment was used for segmentation and image processing tasks. Linear and non linear regressions were used to estimate fish weight. Seven variables extracted from each image (length، height، area، perimeter، equal diameter، major axis length and minor axis length) in four models of mathematical approach (linear، logarithmic، binomial and exponential) were considered for developing each weight prediction equation. Results indicated that fish weight can be estimated with R2 values of 95. 4% and 94% for Tiger-toothed croaker and Silver pomfret، respectively. Model validation was investigated with new data. Results showed that there is not a significant difference between the actual and estimated weight at 5% significance level for all fish species in this study. It was also concluded that the system can accurately measure the length of the fishes using machine vision technology envisaged in this study. The algorithm was also able to sort two species of fishes including Tiger-toothed croaker and Silver pomfret with an accuracy of 100%.

The chemical-morphological relationships in rainbow trout, Oncorhynchus mykiss, was assessed through photography in light room. The image data were transferred to a laptop and analyzed by MATLAB (Matrix Laboratory) version R2009a. The designated algorithm of 7 features, including length, height, area, perimeter, equivalent diameter, major and minor axis length were extracted from the images. Afterward, chemical compositions of the fish (including crude protein, crude fat, moisture content, pH value and leakage), were determined. Pearson's correlation coefficient was used to correlate visual and chemical properties. Factors were assayed with linear, logarithmic, exponential and power method. Results showed that pH, leakage, protein and fat had a good correlation with fish length, and moisture with fish perimeter, in Quadratic equation (R2= 0.993, 0.991, 0.991, 0.989 and 0.992, respectively). In conclusion, this methodology can be applied for automated chemical analysis and on-line monitoring in fish farms.

Crop yield estimation is one of the most important parameters for information and resources management in precision agriculture. This information is employed for optimizing the field inputs for successive cultivations. In the present study، the feasibility of sugar beet yield estimation by means of machine vision was studied. For the field experiments stripped images were taken during the growth season with one month intervals. The image of horizontal view of plants canopy was prepared at the end of each month. At the end of growth season، beet roots were harvested and the correlation between the sugar beet canopy in each month of growth period and corresponding weight of the roots were investigated. Results showed that there was a strong correlation between the beet yield and green surface area of autumn cultivated sugar beets. The highest coefficient of determination was 0. 85 at three months before harvest. In order to assess the accuracy of the final model، the second year of study was performed with the same methodology. The results depicted a strong relationship between the actual and estimated beet weights with R2=0. 94. The model estimated beet yield with about 9 percent relative error. It is concluded that this method has appropriate potential for estimation of sugar beet yield based on band imaging prior to harvest

Local information about tillage intensity and ground residue coverage is useful for policies in agricultural extension، tillage implement design and upgrading management methods. The current methods for assessing crop residue coverage and tillage intensity such as residue weighing methods، line-transect and photo comparison methods are tedious and time-consuming. The present study was devoted to investigate accurate methods for monitoring residue management and tillage practices. The satellite imagery technique was used as a rapid and spatially explicit method for delineating crop residue coverage and as an estimator of conservation tillage adoption and intensity. The potential of multispectral high-spatial resolution WorldView-2 local data was evaluated using the total of eleven satellite spectral indices and Linear Spectral Unmixing Analysis (LSUA). The total of ninety locations was selected for this study and for each location the residue coverage was measured by the image processing method and recorded as ground control. The output of indices and LSUA method were individually correlated to the control and the relevant R2 was calculated. Results indicated that crop residue cover was related to IPVI، RVI1، RVI2 and GNDVI spectral indices and satisfactory correlations were established (0. 74 - 0. 81). The crop residue coverage estimated from the LSUA approach was found to be correlated with the ground residue data (0. 75). Two effective indices named as Infrared Percentage Vegetation Index (IPVI) and Ratio Vegetation Index (RVI) with maximum R2 were considered for classification of tillage intensity. Results indicated that the classification accuracy with IPVI and RVI indices in different conditions varied from 78-100 percent and therefore in good agreement with ground measurement، observations and field records.