Z. Khosrobeygi; Sh. Rafiee; S. S. Mohtasebi; A. Nasiri
Abstract
Introduction Increasing the production efficiency is an important goal in precision farming. The use of precision farming requires a lot of labor work. Also, due to the risk of agricultural operations, it is not recommended to do it directly by humans. Therefore, it is necessary for agricultural operations ...
Read More
Introduction Increasing the production efficiency is an important goal in precision farming. The use of precision farming requires a lot of labor work. Also, due to the risk of agricultural operations, it is not recommended to do it directly by humans. Therefore, it is necessary for agricultural operations to be carried out automatically. For this reason, the application of robotics in agricultural environments, especially in the greenhouse, is increasing. The first step in automatic farming is autonomous navigation. For autonomous navigation, a robot must be the ability to understand its environment and recognize its position. In other words, a robot must be able to create a map of an unknown environment, locate itself on this map and finally plane for the path. This problem is solvable by Simultaneous Localization and Mapping (SLAM). The SLAM problem is a recursive estimation process. In the other words, when a robot moves in an unknown environment, mapping and localization errors increase incrementally. To reduce these two errors, a recursive estimation process is used to solve the SLAM problem. Materials and Methods In this research, two webcams, made by Microsoft Corporation with the resolution of 960×544, are connected to the computer via USB2 in order to produce a stereo parallel camera. For this study, we used a greenhouse that was located the Arak, Iran. Before taking stereo images, a camera path was designed in the greenhouse. This path may be either straight or curved. The designed path was implemented in the greenhouse. The entire path traversed by a stereo camera was 32.7 m and 150 stereo images were taken. Graph-SLAM algorithm was used for Simultaneous Localization and Mapping in the greenhouse. Using the ROS framework, the SLAM algorithm was designed with nodes and network for connecting the nodes. Results and Discussion For evaluation, the stereo camera locations, every step was measured manually and compared with the stereo camera locations that were estimated in the graph-SLAM algorithm. The position error was calculated through the Euclidean distance (DE) between the estimated points and the actual points. The results of this study showed that, the proposed algorithm has an average of error 0.0679412, standard deviation of 0.0456431 and root mean square error (RMSE) of 0.0075569 for camera localization. In this research, only a stereo camera was used to prepare a map of the environment, but other researches have used multiple sensor combinations. Another advantage of this research related to others was created a 3D map (point cloud) of the environment and loop closer detection. In the 3D map, in addition to determining the exact location of the plant, the height of the plant can also be estimated. Plant height estimate is important in some agricultural operations such as spot spray, harvesting and pruning. Conclusion Due to the risk of agricultural activities, the use of robotics is essential. Autonomous navigation is one of the branches of the robotics. For autonomous navigation, a map of environment and localization in this map is need. The purpose of our research was to provide simultaneous localization and mapping (SLAM) in agricultural environments. ROS is a strong framework for solving the SLAM problem. So that, this problem can be solved by combining different nodes in ROS. The method depended only on the information from the stereo camera because stereo camera provided exact distance information. We believe that this study will contribute to the field of autonomous robot applications in agriculture. In future studies, it is possible to use an actual robot in the greenhouse with various sensors for SLAM and path planning.
P. Fayyaz; S. S. Mohtasebi; A. Jafari; A. Masoudi
Abstract
Introduction Essences or essential oils are aromatic compounds that are found in different organs of the plants. Essences can be classified into three groups of natural, synthetic and natural like. Most of the methods that are used to detect and to distinguish essential oils are based on chromatographic ...
Read More
Introduction Essences or essential oils are aromatic compounds that are found in different organs of the plants. Essences can be classified into three groups of natural, synthetic and natural like. Most of the methods that are used to detect and to distinguish essential oils are based on chromatographic methods. However, these analytical methods are time consuming and require expert operators to work with required devices. Moreover, it is necessary to prepare the samples. An electronic nose is known as a tool for mimicking the sense of smell. This tool usually consists of an array of sensors which are used to identify and to isolate a variety of complex odors at a low cost. Since there has been no research on the usage of an electronic nose system for detection and separation of essential oils, the purpose of this study is to develop and to evaluate an electronic nose system for identification and classification of various types of commercial lemon essential oils (synthetic types). Materials and Methods The proposed system consists of a sensor chamber, a sample chamber, an array of MOS sensors, electro valves, a pump, a data acquisition cart and, a processor. Essential oils used in this study includes eight types of synthetic commercial lemon essential oils that were prepared by ((Avishan Khane Tabiat Sabz)) Company located in chemistry and chemical engineering research center of Iran. One gram sample of each essential oil was prepared to be placed in the sample chamber. Each experiment was carried out in 9 replicates and in three stages of 1- Baseline correction (250 s) 2- Sample smell injection (400 s) and 3- Sensors chamber cleaning (200 s). Data received from the sensors signals were initially preprocessed and normalized and then three methods of principal component analyses (PCA), linear discriminant analyses (LDA) and artificial neural network (ANN) were used to process the data. Both PCA and LDA methods were run using the Unscramble x10.4 software and the artificial neural network was used with the help of NeuroSolution 5 software. In ANN, the classification was carried out using a multilayer perceptron (MLP) and Leave-one-out technique. Leave-one-out is an acceptable method for evaluating the performance of the classification algorithms when the number of samples is low. Results and Discussion In order to analyze the data obtained from the sensor array, first, the principal components analysis (PCA) method was used. In this method, the first two principal components of PC 1 and PC 2 totally covered 99% of the data variance. Another plot called as loading plot was used to determine the effects of each sensor responses in pattern recognition analyzes. According to this plot, all sensors had high loading coefficients. In case of distinguishing the lemon essential oils, the results of the linear discriminant analysis (LDA) method showed that this method can distinguish eight types of lemon essential oils with an accuracy of %98. The artificial neural network (ANN) also separated the essential oils with the accuracy of the above %91. Conclusion An Electronic nose system based on semiconductor metal oxide sensors is a powerful tool that can be used as a substitute for traditional methods. In general, this study showed that the electronic nose system based on MOS sensors has the ability to detect and to distinguish commercial lemon essential oils. Considering the wide ranges and economical nature of the essential oils, it is suggested that applications of the electronic nose can be more expanded in the subject of the essential oils of different products.
M. Hajinezhad; S. S. Mohtasebi; M. Ghasemi-Varnamkhasti; M. Aghbashlo
Abstract
Introduction Honey is a supersaturated sugar and viscose solution taken from the nectar of flowers, collected and modified by honeybees. Many producers of honey add some variety of sugars in honey that make difficulties with detection of adulterated and pure honey. Flavor is one of the most important ...
Read More
Introduction Honey is a supersaturated sugar and viscose solution taken from the nectar of flowers, collected and modified by honeybees. Many producers of honey add some variety of sugars in honey that make difficulties with detection of adulterated and pure honey. Flavor is one of the most important parameters in the classification of honey samples and the smell emitted by the honey depending on the different flowers and constituents that could be different. This causes using an electronic nose system to detect honey adulteration. Materials and Methods Honey samples used in this study were lotus honey that was supplied from a market in Karaj city, Alborz province, Iran. Adulterated honey, along with percentages of fraud (by weight) of zero, 20, 35 and 50 percent, was prepared by mixing sugar syrup. Each group of samples, nine times were tested by the electronic nose system. The proposed system, consists of six metal oxide semiconductor sensors, sensor chamber, sample chamber, data acquisition systems, power supply, electric valves, and pumps. Electronic nose is planned for three-phase system baseline correction, the smell of sample injection and cleaning of the sensor and sample chambers with clean air (Oxygen). Responses of the sensors were collected and stored in 420 seconds by a data acquisition system and LabView ver 2012 software. We used fractional method in this study, in order to improve the quality of the information available and to optimize the array output before passing it on to the pattern recognition system. Linear Discriminant Analysis (LDA), Principal Component Analysis (PCA) and Artificial neural network (ANN) were the methods used for analyzing and recognizing pattern of electronic nose signals. Data processing was carried out using Microsoft Excel, neuralsolution 5 and Unscrambler X 10.3 (CAMO AS, Norway). Results and Discussion PCA Results PCA reduces the complexity of the data-set and is performed with no information on the classification of samples. It is based on the variance of the data-set. For PCA analysis, overall PC1 and PC2 explained 91% of the total variance among Lotus honey samples and the adulterations (PC1=80% and PC2=11%). The results indicate that it is clearly possible to recognize Lotus honey with adulterant using electronic nose systems. LDA Results The LDA method for the detection of adulterated honey samples using leave-one-out validation was estimated. The method is most widely used as a method of classification that maximizes variance between the clusters and minimizes variance of within classes. By applying LDA on the collected data, 100% accurate classification for detecting of honey and their adulterations was obtained. It can also be concluded that this method could recognize adulterated honey samples properly. ANN Results The back propagation multilayer perceptron algorithm was used to classify and to detect honey and adulterated types. Performance evaluations of each designed networks were compared by mean square error (MSE) and correlation coefficient (r).The data were divided to three subsets: 60% was used for training, 20% for testing and the remaining 20% were kept for cross validation.After network training and validation using optimized ANN model, i.e. 6-8-4 structure, success rate for 4 outputs (0, 20, 35 and 50% adulterated levels)were found to be 100%.After detecting adulteration, e-nose system accompanied with ANN can accurately classify honey from honey mixtures with fraud materials. Conclusion An electronic nose based on six metal oxide semiconductor sensors was used to detect adulterated honey samples. Electronic nose system can successfully classify between original honey and the adulterated one by pattern recognition method. The PCA, LDA and ANN techniques and analyzes of the electronic nose were very useful for evaluating the quality of the lotus honey. The results of these methods were used to classify the fraud in Lotus honey. However, there is a need to do more researches on the detection of adulteration in other agricultural and food products by electronic nose system.
A. Safrangian; L. Naderloo; H. Javadikia; M. Mostafaei; S. S. Mohtasebi
Abstract
Introduction Vibrations include a wide range of engineering sciences and discuss from different aspects. One of the aspects is related to various types of engines vibrations, which are often used as power sources in agriculture. The created vibrations can cause lack of comfort and reduce effective work ...
Read More
Introduction Vibrations include a wide range of engineering sciences and discuss from different aspects. One of the aspects is related to various types of engines vibrations, which are often used as power sources in agriculture. The created vibrations can cause lack of comfort and reduce effective work and have bad influence on the health and safety. One of the important parameters of the diesel engine that has the ability to create vibration and knocking is the type of fuel. In this study, the effects of different blends of biodiesel, bioethanol and diesel on the engine vibration were investigated. As a result, a blend of fuels such as synthetic fuel that creates less vibration engine can be identified and introduced. Materials and Methods In this study, canola oil and methanol alcohol with purity of 99.99% and the molar ratio of 6:1 and sodium hydroxide catalyst with 1% by weight of oil were used for biodiesel production. Reactor configurations include: maintaining the temperature at 50 ° C, the reaction time of 5 minutes and the intensity of mixing (8000 rpm), and pump flow, 0.83 liters per minute. A Massey Ferguson (MF) 285 tractor with single differential (2WD), built in 2012 at Tractor factory of Iran was used for the experiment. To measure the engine vibration signals, an oscillator with model of VM120 British MONITRAN was used. Vibration signals were measured at three levels of engine speed (2000, 1600, 1000 rpm) in three directions (X, Y, Z). The analysis performed by two methods in this study: statistical data analysis and data analysis using Adaptive neuro-fuzzy inference system (ANFIS). Statistical analysis of data: a factorial experiment of 10×3 based on completely randomized design with three replications was used in each direction of X, Y and Z that conducted separately. Data were compiled and analyzed by SPSS 19 software. Ten levels of fuel were including of biodiesel (5, 15 and 25%) and bioethanol (2, 4 and 6%), and diesel fuel. Data analysis by ANFIS: ANFIS is the combination of fuzzy systems and artificial neural network so that it has both benefits. This system is useful to solve the complex non-linear problems in agricultural engineering applications such as systems involved in the soil, plant and air. ANFIS by linguistic concepts can establish and inference non-linear relationship between inputs and outputs. In this research, modeling was generally performed by Toolbox of ANFIS and coding in MATLAB too. Five important and effective factors in modeling were optimized until the best ANFIS model is obtained. The five factors were: type of input fuzzy sets, the number of input fuzzy sets, fuzzy set of output, methods of optimization and the number of epochs. Results and Discussion Based on the total vibration acceleration values for different fuels in different rpm, pure diesel (B5E4D91) had the highest vibration and the lowest vibration was seen in the mixed fuel of B25E4D71. Based on the results, two combined fuel of (B25E2D73, B25E4D71) have the lowest vibration and highest amount of biodiesel fuel (25%). After them, three combined fuels of (B5E2D83, B5E4D81, and B5E6D79) have created more vibration and the lowest amount of biodiesel fuel in this study (5%) has created the greatest amount of vibration. With increasing engine speed, the number of combustion courses and piston shock per unit of time increases. As a result, the engine body vibration increases. The results are consistent with results from other researchers. Conclusion In this study, motor vibration of MF285 tractors, by replacing a portion of diesel fuel with biodiesel produced from canola oil and bioethanol, was investigated. In the beginning, necessary biodiesel fuel was produced by research reactor in biodiesel workshop, and then different percentages of diesel and bio-ethanol were mixed to biodiesel and ten combined fuels were created. Finally the effect of different fuel combinations and different engine rotational speeds on the tractor engine vibrations was studied based on a factorial randomized complete block design and then analyzed and modeled by ANFIS. The results showed that the vibration of pure diesel fuel had the highest vibration. Also, with increasing biodiesel fuel blends, the amount of vibration reduced significantly. Increase in engine speed had direct effect on increasing the amount of vibration. Also by increasing the percent of bioethanol from 0 to 4%, the amount of vibration was reduced then vibration value increased by raising the percent of bioethanol. After modeling and analyzing, our results showed that the best fuel in terms of having the lowest vibration motor was B25E4D71.
A. Sanaeifar; S. S. Mohtasebi; M. Ghasemi-Varnamkhasti; H. Ahmadi
Abstract
Aroma is one of the most important sensory properties of fruits and is particularly sensitive to the changes in fruit compounds. Gases involved in aroma of fruits are produced from the metabolic activities during ripening, harvest, post-harvest and storage stages. Therefore, the emitted aroma of fruits ...
Read More
Aroma is one of the most important sensory properties of fruits and is particularly sensitive to the changes in fruit compounds. Gases involved in aroma of fruits are produced from the metabolic activities during ripening, harvest, post-harvest and storage stages. Therefore, the emitted aroma of fruits changes during the shelf-life period. The electronic nose (machine olfaction) would simulate the human sense of smell to identify and realize the complex aromas by using an array of chemical sensors. In this research, a low cost electronic nose based on six metal oxide semiconductor (MOS) sensors were designed, developed and implemented and its ability for monitoring changes in aroma fingerprint during ripening of banana was studied. The main components are used in the e-nose system include sampling system, an array of gas sensors, data acquisition system and an appropriate pattern recognition algorithm. Linear Discriminant Analysis (LDA) technique was used for classification of the extracted features of e-nose signals. Based on the results, the classification accuracy of 97/3% was obtained. Results showed the high ability of e-nose for distinguishing between the stages of ripening. It is concluded that the system can be considered as a nondestructive tool for quality control during banana shelf-life.
