---

Some physical attributes of two common types of Iranian garlic cloves (white and pink) were identified and compared. A machine vision system was used to determine three di-mensions and both major and minor projected areas of garlic cloves at a moisture content of 42.4% w. b. The geometric mean diameter and sphericity were calculated, as well as the unit mass and volume of cloves were measured. In the moisture range from 34.9 to 56.7% w.b., the unit density, bulk density and porosity for both types were measured. Re-sults showed that the unit density, bulk density and porosity of cloves were affected sig-nificantly by moisture content (p<0.01). The type of garlic had a highly significant effect on the unit density and porosity (P<0.01), and a significant effect on the bulk density (P<0.05). The relationship between volume and dimensions of cloves was established using regression analysis. The effect of moisture content on physical properties of cloves was also expressed by appropriate equations.

---

---

Apple fruit as horticultural products is considered very valuable in terms of food production and employment and exchange technology in Iran. Due to increasing industrial production methods and significant public demand of dried apple layer as an apple product, qualitative separation methods for long time lasting of this product becomes more important. In microscopic scale of fast chilling method, internal surface of dried apple layers consist of Crystal-like beads which are different in size and shape. Finally, arrangements make unique structure before and after of reduction lasting quality. In this research, a technique was presented to fast sorting during constant time process using machine vision technology. At first, the encryption operation for identification of Defined parameters using two methods of Wavelet decomposition and Wavelet packets was done and finally pictures sorting for identification of dried desirable layers from inferior layers using energy values in images after encryption operation were done. Results showed that the Wavelet packets method was more capable in images encryption than the other method. Energy interval detection threshold in images was found quite effective which can identify images of 2600×2600 pixels desirable layers in 0.86 second.  

---

  Apple fruit is one of the most worthy garden Product with high nutritional Value and its production in Iran makes more job and Exchange technology. From different apple Non-destructive quality control methods, machine vision technology achieves the more speed, quality, greater productivity and higher valuation for the product. Usually, apple bruise overlaps with Peduncle and in these causes, serious problems of recognition for quality sorting occurs. In this research work it was tried to work out this problem and to increase the sorting systems performance precision. In order to accomplish this, two separate algorithms based on color to identify bruise and pedicle was designed in Matlab. It was achieved 97.14% accuracy for the bruise algorithm and 100% accuracy for the pedicle algorithm. Then with integration of these two algorithms, an algorithm was achieved with 94.29% accuracy. Further experiments to investigate the possibility of increasing the accuracy in detecting bruise with time maintenance was performed by the bruise algorithm. The results indicate that the bruise detection quality by this algorithm gradually increased and after two to three days it reaches the desired consistency. Another algorithm with special properties of bruise and pedicle pictures shape such as roundness value, ratio of area to Perimeter square and also coefficient of variation (cv) of distances of spaced points on the edge from center of gravity of picture was designed. Then bruise and pedicle were distinguished from each other with an accuracy of 100% with this algorithm along with the ANN which it proving the importance of using these techniques, combined with machine vision techniques to increase the accuracy of sorting machines performance.  

---

The diversity and abundance of quality characteristics of agricultural products, has been the main reason for the development of non-destructive methods. Machine vision and artificial intelligence are powerful techniques for diagnosing most physical, mechanical and chemical properties of agricultural products. Before export fruits are classified by shape, volume and weight. Ranking fruit through taste (sweet or tart) non-destructively plays an important role in marketing, choice power and its application. In this research, it was detect the taste of Thompson orange while combining artificial intelligence (AI) and visual machine technique. A closed circuit digital installed in special frame, under specific height and light was used to take picture from samples vertically. Also, an algorithm (program) based on AI was developed to diagnose the variety and taste of Thompson orange through apparent characteristics in Matlab software. The results showed that the success rate of taste determination for Thompson orange using ANFIS and ANN-GA (Artificial Neural Network-Genetic Algorithm) was 96.67 and 90.0% respectively.  

---

A method based on Machine Vision System (MVS) is hereby employed to evaluate grape drying through an assessment of the fruit’s shrinkage and quality during the dehydration. Experimental data as well as captured images are obtained at an air velocity of 1.4 m s^-1 and different drying temperatures (50, 60, 70ºC). The results indicated the effect of temperature on the moisture content, shrinkage and color changes. The moisture content along with color changes (ΔE) were modeled and linear regressions applied to correlate the fruit’s shrinkage as well as color features to the moisture content. The results obtained, displayed that there existed good linear relationships between the fruit’s moisture content, and shrinkage as well as color. The results also revealed that the moisture content vs. quality of the grape could be online evaluated through machine vision during the drying process.

---

The Modal analysis is one of the applicable methods used to identify the dynamic characteristics of structures. Inspection of structures to avoid resonance conditions can be achieved by extracting vibration modes using modal analysis. Since every point of the vibrating structure has its own characteristics such as the displacement, speed and acceleration, therefore the measurement of these parameters in a specific time interval can be used to extract modal parameters. In this study, stereo vision as a non-contact measuring system is used to obtain the displacement of several points of the blade of a 2.5kW wind turbine with a length of 3m under the operational modal condition. At first, the camera calibration process is performed and then the three-dimensional data of the turbine blade are extracted from images recorded during the test. Consequently, modal parameters of the blade are calculated by analyzing the data. Finally, modal parameters obtained by three different methods including the stereo vision system, the finite element analysis and the testing accelerometer are compared. The results show that visually obtained data are sufficiently accurate to find the natural frequency of the first mode of the blade. The first natural frequency mode extracted by the stereo vision System shows a difference of 10.36% and 2.67% compared to the those obtained by finite element method and the accelerometer respectively.

---

This paper focuses on the problem of finding object orientation around Yaw & Pitch & Roll angels. The object orientation is computed in a real time manner using a mono-camera and three points on a solid object in a machine vision software. Three points should be selected from environment at the beginning. In order to reduce wreckful effects of environmental lights on detecting colorful objects and also to reduce the number of used software filters, IR LEDs with 850nm invisible wavelength are used. Artificial Neural Network (ANN) is used for solving this problem since orientation's equations are nonlinear and real-time solving for them is impossible. For solving the problem a feed forward artificial neural network with one hidden layer and 21 nodes in that is used, which has 3 nodes for output layer and 6 nodes for input layer. For having high accuracy in ANN, output data is also obtained from a MPU-9150 installed on a 2-DOF orientional parallel robot and compared to ANN outputs. 7243 data from Roll and Yaw angles and 751 data from Pitch angle is obtained from MPU-9150 sensor and the later 2-DOF orientional parallel robot and 467 data remains nonuse for learning ANN. After learning the neural network, results compared to nonuse data for ANN learning and desire results obtained with 0.038 maximum error

---

This paper presents a new method based on machine vision for mobile robots to detect and avoid obstacles in unknown environments. One of the challenges of mobile robots trajectory control in unknown environments is that their obstacle avoidance system to be designed robust to material and shape of the obstacle. In this research a mobile robot equipped with a camera is designed and fabricated. Also an algorithm is proposed and implemented on the robot in order to detect obstacles by image processing and to control the robot trajectory. Three color laser pointers are mounted on the robot with certain angles that emit beams to the ground at ahead of the robot. The received images from camera contain these colored points that their coordinates are determined by image processing. Then position of any possible obstacle is detected using the proposed algorithm and the robot is commanded to avoid obstacles by changing its path. These obstacles can be static or dynamic. Our experimental results show that the proposed method, with a high reliability, has the ability to detect and avoid obstacles with any shape and material whereas other similar methods had restrictions in this regard.