This research paper is an attempt to present Content Based Image Retrieval (CBIR) system developed for retrieving diseased leaves of soybean. It uses color, shape and texture features of leaf. Color features are extracted using HSV color histogram. Scale Invariant Feature Transform (SIFT) provides shape features in the form of matching key points. Local Binary Pattern (LBP) and Gabor filter are widely used texture features. Novel texture feature named Local Gray Gabor Pattern (LGGP) is proposed by combining LBP and Gabor. Performance of all these features with respect to retrieval precision is tested for three soybean leaf diseases. Further color, shape and texture features are combined to increase performance. It is found that when LGGP is combined with color histogram and SIFT retrieval precision is improved. Retrieval efficiency of about 96%, 68% and 76% is achieved for soybean leaves affected by mosaic virus, septoria brown spot and pod mottle disease respectively. Average retrieval efficiency of 80% (for the top 5 retrieval) and 72% (for the top 10 retrieval) is obtained by combined features. This retrieval precision is database dependent and varies with size of the database and quality of images.

View full abstract

Because of uncertain nature of sensory evaluation due to differences in the individual panelist's perception of the product attributes, application of fuzzy set concept could be useful. In this research, adaptive neuro-fuzzy inference system (ANFIS) was used to predict overall acceptability of ice cream. Consumer acceptance has been recognized as the key driver for product process. Experimental sensory attributes (flavor, body & texture, viscosity and smoothness) were used as inputs and independent overall acceptability as output of ANFIS. Thirty percent, thirty percent and forty percent of the sensory attributes data were used for training, checking and testing of the ANFIS model, respectively. It was found that ANFIS model achieved an average prediction error of overall acceptability of ice cream of only 5.11%. These results indicate that this model could potentially be used to estimate overall sensory acceptance of ice cream and related products.

View full abstract

In this study, a visual-based furrow line detection method was developed for navigating an autonomous robot vehicle in an agricultural field. The furrow line detection method integrates a crop or non-crop field identification method, two types of box filters, which are a color-based furrow detection filter and a grayscale separability-based furrow detection filter, and a robust furrow line parameter estimator. In experiments, the performance of our developed method was tested on more than 8000 images of 17 types of test fields: nine types of crop fields (sweet pea, green pea, snow pea, lettuce, Chinese cabbage, cabbage, green pepper, tomato, and tea), and eight types of tilled soil fields. By using a wide camera angle with a low depression angle, the detection rate of the furrow line was 98.0%, the root mean square error (RMSE) of positioning of the furrow line was 12.1 pixels, and the RMSE of angle of the furrow line was 3.8°. Moreover, by using the oblique camera angle, the detection rate was 93.4%, the RMSE of positioning was 23.3 pixels, and the RMSE of angle was 6.1°. The results showed that our method using the wide and oblique camera angles could approximately detect the furrow line in the test fields. The average processing speed was approximately 2.5 Hz for the crop fields and 4.0 Hz for the tilled soil fields. Our method demonstrated a high potential to robustly and precisely detect a single targeted furrow line in the 17 types of test fields.

View full abstract

The assessment of plant water status under drought conditions is essential for the understanding of the adaptation physiological mechanisms of plants, because water stress is one of the most common limitations of crop growth and yield. This study was carried out to estimate the leaf water content in sunflower plants based on spectral reflectance in the near-infrared region. Two varieties of sunflower plants were cultivated during two autumn seasons inside a non-acclimatized greenhouse until the start of the flowering stage, and later were maintained in a growth chamber with the purpose of submitting the plants to a slow and progressive dehydration rate during 12 consecutive days. Measurements of spectral reflectance and leaf mass (fresh and dry) were accomplished along the water stress period. Increases on the spectral reflectance values were observed as the leaf water loss was intensified. The coefficients of determination (R²) of models between leaf water content (LWC) and water index (WI) were 0.948 and 0.956 for Sunbright and Sunbright Supreme varieties, respectively, while the R² between LWC and wavelength where the first-order derivative of the spectral reflectance is minimum (λd_min) were 0.979 and 0.988 for Sunbright and Sunbright Supreme varieties, respectively. Based on the results, the radiometric indicators (WI andλd_min) might be useful tools in the development of automatic systems, resulting in a non-destructive, simple, easy, and instantaneous method for monitoring water status in sunflower plants.

View full abstract

A simulation approach based on transport phenomena is proposed in this paper, to help manage the logistics of fresh broccoli during cold chain. Indeed, for many varieties of fresh vegetables their shelf-life depends critically on storage temperature progress, affecting the visual quality (VQ) as a consequence of changes in intrinsic quality parameters. Broccoli was chosen as a model food, and average ammonium (NH₄⁺) concentration was adopted as a proper quality parameter associated with residual VQ. Therefore, a mathematical model for predicting the residual VQ is proposed depending on the keeping temperature (among 5 and 20 °C), for a relatively long storage duration (up to 12 days). Model validation is brought over by comparing experimental and computed VQ.
The predictions were affected by maximum errors of about 3%, 13% and 16% for runs at 5, 10 and 20 °C respectively. The effect of temperature fluctuations were then explored. Two non-isothermal scenarios commonly occurring during retail stores in European and US handlers were first analyzed to test the model's ability to predict the visual quality score. Moreover, the model can be easily applied to assess the household or retail shelf-life, or in general the expiration date after a given storage duration.

View full abstract

Visible transmission spectroscopy combined with multivariate analysis was used to develop a nondestructive detection system for hatching egg fertility. The transmission spectra (500–750 nm) of 165 light brown-shell broiler hatching eggs were acquired between incubation time 0–144 h at 24 h interval. Several multivariate classification methods, i.e. k-means clustering, linear discriminant analysis (LDA) and support vector machine (SVM) were used to discriminate between fertile and infertile eggs. The LDA and SVM classification models achieved 100% classification accuracy for both fertile and infertile eggs, while the k-means clustering model had a classification accuracy of 96% for fertile eggs and 100% for infertile eggs at 96 h of incubation. These results demonstrate that visible transmission spectroscopy, combined with an appropriate multivariate method, has the potential to use for detection of infertile eggs in commercial hatchery operations.

View full abstract

A program in visual basic software was developed for predicting tractor implement performance and energy mapping of tractor implement combination. A simple and complete instrumentation system for evaluation of tractor performance under actual field condition is discussed. The instrumentation system of tractor was tested under actual field condition for validation of the developed software, and found good results. The software was developed by considering various traction models, fuel consumption models and draft models to fit for most commonly used agricultural tractor implement combinations. The developed software is user friendly and efficient in predicting. This could help the researchers and engineers for power and energy mapping of new tractor and implement models.

View full abstract

Organic manures play an important role in improvement of soil fertility, increasing of crop yield and quality. The beneficial advantages of manure depend firstly on the method of spreading. Usually current manure spreaders cannot broadcast manures with different moisture contents and have non-uniform application rate. So the main objective of this research is performance evaluation of a new helical two-sided manure distributor that would be capable to spread manure at different moisture contents with long distribution width in uniform patterns. Experimental evaluation of the machine performance at two moisture content of manure in three forward velocities and three rotational speeds of drum showed that the moisture content of manure and rotational speed of drums had significantly effects (p = 0.01) on the distribution width and uniformity of the distribution pattern. The maximum distribution width of the machine at moisture content of 23% w.b. and 54% w.b. and in 1000 rpm rotational speed of drum were 19.3 m and 23.3 m respectively. Results indicated that the manure distribution uniformity increased by increasing the manure moisture and the distribution pattern of machine has an acceptable correlation with oval distribution pattern (p = 0.05).

View full abstract

A machine vision system with double lighting and an image processing algorithm were developed to detect undesirable objects in the paddy as it is being harvested to provide information for adjustment of parameter settings by the operator and automate combine harvester procedures. The system consisted of a web camera, a double lighting system (frontlight and backlight), an image acquisition, and image analysis algorithm. Images captured by this system were used for detection of undesirable objects (long rachis branches, grass and leaves, straw and stalks) and damaged grain (brown rice and cracked rice). Evaluation of the results demonstrate the proposed detection algorithm had a detection correlation of determination, R² greater than 0.70 for detection of each of the undesirable objects. Future application of this system in the field could help the operator of the combine harvester to improve the efficiency and adjustment of automated harvesting procedures, as well as the quality of the harvested paddy.

View full abstract

A load control system was designed to keep the threshing load of the small-sized harvester in the rated load range, and the current detection for the motor driving the harvester cylinder was presented as a new method to detect the cylinder torque indirectly. First the mathematical relationship was established between the cylinder torque and the motor current, and the cylinder-load model was presented. Then the motor current according to the rated cylinder load was obtained by tests as the reference for the load control system. The results showed that the coefficient of variation of consistency of the motor current decreased from 10.7% by manual control to 7.4% by the load control system, and labor intensity index decreased to 8 from 13 after using the load control system. It indicates that the cylinder load control can improve the harvester performance and relieve manual labor required.

View full abstract

This paper presents a discrete-time greenhouse climate model to meet the requirement of model-based greenhouse climate discrete control. In this model, three state variables including the indoor temperature, humidity and CO₂ concentration and several control actions are considered. To save several states, algebraic fitting technique and time sequence functions are used to estimate some variables such as the temperatures of cover, canopy and floor. To improve simulation performance of the discrete-time model, polynomial is employed to approximate some unmodeled dynamics, and least square technique and Levenberg-Marquardt (LM) algorithm are used to identify the parameters of relevant terms in the model. The model is validated by a complete greenhouse climate model and a Venlo type commercial greenhouse, and the results indicate that the proposed model demonstrates good simulation performance, and is suitable to be used to greenhouse climate control design.

View full abstract