Browsing by Author "Kurtulmuş, Ferhat"

Now showing 1 - 20 of 21

Aerodynamic performance of some wind turbine rotor models
(Uludağ Üniversitesi, 2010) Vardar, Ali; İzli, Nazmi; Kurtulmuş, Ferhat; Uludağ Üniversitesi/Ziraat Fakültesi/Tarım Makinaları Bölümü.
In this study, power coefficients of miniature wind turbine rotors produced by using NACA profiles were determined. To this end, an open system for wind turbine rotor model was established. In this test system power values of the wind that reach to each of the rotor form and tip speed ratios of each rotor form were indicated. With the help of these data, power coefficients of each rotor form were calculated. It was found that power coefficient values that were obtained in the study changed between 0,316 and 0,416. It was determined that 4-bladed rotors that were produced by using NACA 4415 and NACA 5317 profiles, having 0 degree twisting angle and approximately 10 degrees binding angle, had the highest ideal power coefficient values.
Bilgisayarlı görme esaslı değişken oranlı bir alev makinası için görüntü alma sisteminin optimizasyonu
(Bursa Uludağ Üniversitesi, 2020-01-21) Kargacı, Kübra; Kurtulmuş, Ferhat; Sefil, Kadir Tayfun; Arslan, Selçuk; Bursa Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.
Bu çalışmanın amacı, yabancı ot kontrolü için kullanılan bir alev makinesi prototipine entegre edilecek görüntü işleme esaslı, yapay aydınlatmalı bir görüntü alma sistemi geliştirmektir. Ayrıca, düşük maliyetli bir gömülü devre ve kamera (Raspberry Pi 3) kullanan görüntü işleme sisteminin yapay aydınlatmalı görüntü alma odacığının gerekli işletme parametrelerini belirlemek hedeflenmiştir. Görüntü işleme algoritmalarının geliştirilmesi ve sayısal analiz aşamalarında OpenCV açık kaynak kodlu görüntü işleme kütüphanesi ve Python programlama dili kullanılmıştır. Sistem geliştirme aşamalarında geliç (SorghumhalepenseL.), pıtrak (Xanthiumstrumarium L.), tarla sarmaşığı (Convolvulusarvensis L.) ve köygöçüren (Cirsiumarvense) otlarının bulunduğu bir arazi koşulunda yabancı otların görüntüleri alınmıştır. Dış ortama açık ve yapay aydınlatma sistemiyle alınan görüntülerin histogramları karşılaştırılmıştır. Yabancı ot piksel dağılımları incelenerek ikilileştirme için uygun eşik değerleri belirlenmiştir. Sonuç olarak, geliştirilen algoritma 20 fps’ye yakın hızlarda hareketli görüntüler üzerinde çalıştırılarak anlık yabancı ot oranlarının belirlenebileceği bulunmuştur. Geliştirilen sistem kullanılarak test görüntülerinde yabancı ot piksel oranı %98’lik bir başarı ile hesaplanabilmiştir.
Classification of chestnuts according to moisture levels using impact sound analysis and machine learning
(Springer, 2018-12) Kavdir, İsmail; Kurtulmuş, Ferhat; Öztüfekçi, Sencer; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; Uludağ Üniversitesi/Ziraat Fakültesi/Toprak Bilimi ve Bitki Besleme Bölümü.; R-8053-2016; 15848202900; 57189374728
In this study, a prototype system was designed, built and tested to classify chestnuts using impact sound signals and machine learning methods according to moisture contents. Briefly, the system consisted of a shotgun microphone, a sliding platform, an impact surface, a triggering system, a sound device and a computer. Sound signal data were acquired from 2028 chestnut samples with three different moisture levels. Acoustic signals from chestnut samples were filtered to alleviate negative effects of unwanted noise. Four machine learning classifiers using three different feature sets obtained from two feature groups applying feature reduction methods were trained and tested to classify pairs of chestnut moisture group categories as 35% versus 45%, 35% versus 55%, 45% versus 55% (classification with two outputs) and 35% versus 45% versus 55% (classification with three outputs), respectively. The highest classification success (88%) was achieved for the classification application category of 35 versus 55%.
Classification of olives using FT-NIR spectroscopy, neural networks and statistical classifiers
(Springer, 2018-12) Kavdır, İsmail; Büyükcan, Burak M.; Kurtulmuş, Ferhat; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; R-8053-2016; 15848202900
Green olives (Olea europaea L. cv. Ayvalik') were classified based on their surface features such as existence of bruise and fly-defect using two NIR spectrometer readings of reflectance and transmittance, and classifiers such as artificial neural networks (ANN) and statistical (Ident and Cluster). Spectral readings were performed in the ranges of 780-2500 and 800-1725nm for reflectance and transmittance modes, respectively. Original spectral readings were used as input features to the classifiers. Diameter correction was applied on reflectance spectra used in ANN classifier expecting improved classification results. ANN classifier performed better in general compared to statistical classifiers. Classification performance in detecting bruised olives using diameter corrected reflectance features and ANN classifier was 99% while it was 98% for Ident and Cluster classification approaches using regular reflectance features. Classification between solid and fly-defected olives was performed with success rates of 93% using reflectance features and 58% using transmittance features with ANN classifier while statistical classifiers of Ident and Cluster performed between 52 and 78% success rates using the same spectral readings. ANN classifier resulted 92% classification success for the classification application considering three output classes of solid, bruised and fly-defected olives using reflectance features while it performed 57.3% success rate using transmittance features.
Classification of pepper seeds using machine vision based on neural network
(Chinese Acad Agricultural Engineering, 2015-12-22) Kavdir, İsmail; Kurtulmuş, Ferhat; Alibaş, İlknur; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; 0000-0002-1898-8390; AAH-4263-2021; R-8053-2016
Pepper is widely planted and used all over the world as fresh vegetable and spice. Genetic and morphological information of pepper are stored through seeds. Determination of seed variety is crucial for correctly identifying genetic materials. Pepper varieties cannot be easily classified even by an expert eye due to the very small size of seeds and visual similarities. Hence, more advanced technologies are required to determine the variety of a pepper seed. A classification method was proposed to discriminate pepper seed based on neural networks and computer vision. Image acquisition was conducted using an office scanner at a resolution of 1200 dpi. Image features representing color, shape, and texture were extracted and used to classify pepper seeds. By calculating features from different color components, a feature database was constructed. Effective features were selected using sequential feature selection with different criterion functions. As a result of the feature selection procedure, the number of the features was significantly reduced from 257 to 10. Cross validation rules were applied to obtain a reliable classification model by preventing overfitting. Different numbers of neurons in the hidden layer and various training algorithms were investigated to determine the best multilayer perceptron model. The best classification performance was obtained using 30 neurons in the hidden layer of the network. With this network, an accuracy rate of 84.94% was achieved using the sequential feature selection and the training algorithm of resilient back propagation in classifying eight pepper seed varieties.
Detecting corn tassels using computer vision and support vector machines
(Pergamon-Elsevier Science Ltd, 2014-11-15) Kavdır, İsmail; Kurtulmuş, Ferhat; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; R-8053-2016; 15848202900
An automated solution for maize detasseling is very important for maize growers who want to reduce production costs. Quality assurance of maize requires constantly monitoring production fields to ensure that only hybrid seed is produced. To achieve this cross-pollination, tassels of female plants have to be removed for ensuring all the pollen for producing the seed crop comes from the male rows. This removal process is called detasseling. Computer vision methods could help positioning the cutting locations of tassels to achieve a more precise detasseling process in a row. In this study, a computer vision algorithm was developed to detect cutting locations of corn tassels in natural outdoor maize canopy using conventional color images and computer vision with a minimum number of false positives. Proposed algorithm used color informations with a support vector classifier for image binarization. A number of morphological operations were implemented to determine potential tassel locations. Shape and texture features were used to reduce false positives. A hierarchical clustering method was utilized to merge multiple detections for the same tassel and to determine the final locations of tassels. Proposed algorithm performed with a correct detection rate of 81.6% for the test set. Detection of maize tassels in natural canopy images is a quite difficult task due to various backgrounds, different illuminations, occlusions, shadowed regions, and color similarities. The results of the study indicated that detecting cut location of corn tassels is feasible using regular color images
Detection of dead entomopathogenic nematodes in microscope images using computer vision
(Academic Press Inc Elsevier Science, 2013-11-05) Kurtulmuş, Ferhat; Ulu, Tufan Can; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; Uludağ Üniversitesi/Ziraat Fakültesi/Bitki Koruma Bölümü Bölümü.; 0000-0003-3640-1474; R-8053-2016; B-6308-2011; 15848202900; 55955615200
Entomopathogenic nematodes are soil-dwelling living organisms which have been widely used for controlling agricultural insect pests as part of biological control. Because easy to use procedures have been developed for their application using standard sprayers, they are one of the best alternatives to pesticides. In laboratory procedures, counting is the most common, laborious, time-consuming and approximate part of the studies conducted on entomopathogenic nematodes. Here, a novel method was proposed to detect and count dead Heterorhabditis bacteriophora nematodes from microscope images using computer vision. The proposed method consisted of three main algorithm steps: pre-processing to obtain the medial axes of the nematode worms as accurately as possible, separation of overlapped nematode worms with a skeleton analysis; and detection of dead nematodes using two different straighter line detection methods. The proposed method was tested on 68 microscope images which included 935 live worms and 780 dead worms. Proposed method was able to detect the worms in microscope images successfully with recognition rates of over 85%. (C) 2013 IAgrE. Published by Elsevier Ltd. All rights reserved.
Determination of pipe diameters for pressurized irrigation systems using linear programming and artificial neural networks
(Ankara Üniversitesi, 2023-01-01) Kurtulmuş, Ezgi; Kurtulmuş, Ferhat; Kuşcu, Hayrettin; Arslan, Bilge; Demir, Ali Osman; KURTULMUŞ, EZGİ; KURTULMUŞ, FERHAT; KUŞÇU, HAYRETTİN; ARSLAN, BİLGE; DEMİR, ALİ OSMAN; Bursa Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; 0000-0001-9600-7685; AAH-4682-2021; AAH-2936-2021; R-8053-2016; JOP-8553-2023; JLX-2232-2023
Pressurized irrigation systems are widespread among other alternatives in Mediterranean countries. Since the initial investment costs of pressurized irrigation systems are quite high, it is crucial to determine design parameters such as pipe diameter. Most of the current optimization techniques for pipe diameter selection are based on linear, non-linear, and dynamic programming models. The ultimate aim of these techniques is to produce solutions to problems with less cost and computation time. In this study, a novel approach for determining pipe diameter was proposed using Artificial Neural Networks (ANN) as an alternative to existing models. For this purpose, three pressurized irrigation systems were investigated. Different ANN architectures were created and tested using hydrant level parameters of the irrigation systems, such as irrigated area per hydrant, hydrant discharge, pipe length, and hydrant elevation. Different training algorithms, transfer functions, and hidden neuron numbers were tried to determine the best ANN model for each irrigation system. Using multilayer feed-forward ANN architecture, the highest coefficients of determination were found to be 0.97, 0.93, and 0.83 for irrigation systems investigated. It was concluded that pipe diameters could be determined by using artificial neural networks in the planning of pressurized irrigation systems.
Developing wind-concentrator systems for the use of wind turbines in areas with low wind-speed potentials
(Wiley-V C H Verlag Gmbh, 2015-12-01) Vardar, Ali; Eker, Bülent; Kurtulmuş, Ferhat; Taşkın, Onur; VARDAR, ALİ; KURTULMUŞ, FERHAT; TAŞKIN, ONUR; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü/Tarımsal Enerji Sistemleri.; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü/Tarım Makina Sistemleri.; 0000-0001-6349-9687; 0000-0002-5741-8841; AAH-5008-2021; AAH-5018-2021; R-8053-2016
The ability to supply energy in rural areas and in agricultural plants with renewable energy technologies, especially wind energy, is advantageous in terms of a sustainable environment and the increasing cost of energy. Today, wind turbines are used actively in many areas, some of which are for commercial purposes. Small-scale wind turbines that produce electricity directly have the necessary characteristics for use in agricultural plants. In this study, wind-concentrator systems for small-scale wind turbines that can be used in agricultural electrification applications have been designed for geographical areas where the wind-speed potential is low. Three different concentrator systems have been designed to make use of low wind-speed potentials and obtain high power values with relatively small rotor diameters. The three different designs have been produced as prototypes, and power values of 324-503 Wm(-2) (at 5 ms(-1) wind speed) can be obtained by concentrating the wind. The efficiency, power, energy production capacity, and economic elements of the models were determined, and the possible results for agricultural plants have been assessed. According to these assessments, the efficiency values are 71 and 90% for wind speed and 410 and 600% for wind power. The energy production capacities are a maximum of 6462, 5193, and 8226 kWhm(-2) per year for the conical wind-concentrator system, the wind-concentrator system with a panel, and the wind-concentrator system without a panel, respectively. If the energy production cost per unit of these systems is considered, these systems are not economical. Therefore, these systems must be produced on a large scale to become economical, and their size must be enlarged to reduce the cost. Consequently, the potential power values per unit area and the potential energy values per unit produced by the wind-concentrator systems will contribute to the production of more energy than that achieved by current wind turbines.
Discriminating drying method of tarhana using computer vision
(Wiley, 2014-03-19) Deǧirmencioǧlu, Nurcan; Kurtulmuş, Ferhat; Gürbüz, Ozan; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; Uludağ Üniversitesi/Ziraat Fakültesi/Gıda Mühendisliği Bölümü.; 0000-0001-7871-1628; R-8053-2016; K-1499-2019; 15848202900; 8528582100
Tarhana is a traditionally fermented wheat flour product of Turkey which has high nutritional value. A rapid and objective evaluation of tarhana quality by assessing the used drying method is important for producers and packaging companies. A computer vision method was developed to discriminate between drying methods of tarhana. Tarhana samples were prepared with three drying methods: sun dried, oven dried and microwave dried. An image acquisition station was constituted under artificial illumination. Different types of machine learning methods and feature selection methods were tested to find an effective system for the discrimination between drying methods of tarhana using visual texture features with different color components. Experimental results showed that the best accuracy rate (99.5%) was achieved with a K-nearest-neighbors classifier through the feature model based on stepwise discriminant analysis.
Discriminating rapeseed varieties using computer vision and machine learning
(Pergamon-Elsevier, 2015-03) Kurtulmuş, Ferhat; Ünal, Halil; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; R-8053-2016; AAH-4410-2021; 15848202900; 55807866400
Rapeseed is widely cultivated throughout the world for the production of animal feed, vegetable fat for human consumption, and biodiesel. Since the seeds are evaluated in many areas for sowing and oilseed processing, they must be identified quickly and accurately for selection of a correct variety. An affordable method based on computer vision and machine learning was proposed to classify the seven rapeseed varieties. Different types of feature sets, feature models, and machine learning classifiers were investigated to obtain the best predictive model for rapeseed classification. The training and test sets were used to tune the model parameters during the training epochs by varying the complexity of the predictive models with grid-search and K-fold cross validation. After obtaining optimized models for each level of complexity, a dedicated validation set was used to validate predictive models. The developed computer vision system provided an overall accuracy rate of 99.24% for the best predictive model in discriminating rapeseed variety.
Green citrus detection using 'eigenfruit', color and circular Gabor texture features under natural outdoor conditions
(Elsevier, 2011-09) Lee, Won Suk; Kurtulmuş, Ferhat; Vardar, Ali; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; AAH-5008-2021; R-8053-2016; 15848202900; 15049958800
A machine vision algorithm was developed to detect and count immature green citrus fruits in natural canopies using color images. A total of 96 images were acquired in October 2010 from an experimental citrus grove in the University of Florida, Gainesville, Florida. Thirty-two of the total 96 images were selected randomly and used for training the algorithm, and 64 images were used for validation. Color, circular Gabor texture analysis and a novel 'eigenfruit' approach (inspired by the 'eigenface' face detection and recognition method) were used for green citrus detection. A shifting sub-window at three different scales was used to scan the entire image for finding the green fruits. Each sub-window was classified three times by eigenfruit approach using intensity component, eigenfruit approach using saturation component, and circular Gabor texture. Majority voting was performed to determine the results of the sub-window classifiers. Blob analysis was performed to merge multiple detections for the same fruit. For the validation set, 75.3% of the actual fruits were successfully detected using the proposed algorithm.
Güneşin konumuna göre iki eksende hareket eden sensörsüz bir sistem üzerindeki ışınım ölçer cihazının performansı üzerine bir araştırma
(Bursa Uludağ Üniversitesi, 2020) Dede, Kübra; Kurtulmuş, Ferhat; Bursa Uludağ Üniversitesi/Fen Bilimleri Enstitüsü/Biyosistem Mühendisliği Anabilim Dalı.; 0000-0002-1459-3869
Dünya nüfusu ve tüketimin artmasıyla enerji ihtiyacının karşılanması önemli bir sorun haline gelmiştir. Fosil yakıtların zaman içinde tükenmeleri, çevreyi kirletmeleri ve sera etkisi yaratarak iklim değişikliğine sebep olmaları gibi nedenler güneş, rüzgâr, jeotermal ve biyokütle gibi yenilenebilir enerji kaynaklarının yaygın olarak kullanılmaya başlanmasını sağlamıştır. Fosil yakıtlar yerine yenilenebilir enerji kaynaklarının aktif olarak kullanılmasını sağlamak amacıyla her geçen gün yeni teknolojiler geliştirilmektedir. Söz konusu teknoloji ürünlerinin performanslarının ölçümlenebilir olması yenilenebilir kaynaklardan optimum miktarda verim elde edilebilmesi bakımından önemlidir. Örneğin bir güneş paneli vasıtasıyla güneşten enerji elde edilirken panellerin güneşe dik açı ile konumlandırılarak güneş ışınlarını dik açı ile almaları sağlanmaktadır. Panellerin güneşe olan açıları farklı yöntemlerle tespit edilerek en uygun konum bulunabilmektedir. Bu çalışmada güneş ışınımının ölçülmesinde, manuel olarak el ile nişan alma yöntemi yerine optik izleme sönsörü kullanılmaksızın konum ve zamana bağlı olarak güneş açılarının hesaplanmasıyla iki eksende hareket eden bir güneş takip sistemi cihazı geliştirmek amaçlanmıştır. Bu takip sistemi üzerine yerleştirilen ışınım ölçer cihazının performansında nasıl bir değişim olacağı gözlemlenmiştir. Aynı konum ve saatte manuel olarak çalıştırılan bir ışınım ölçer cihazından elde edilen veriler ile geliştirilen takip sistemine bağlı diğer ışınım ölçer cihazından alınan veriler karşılaştırılmıştır. Karşılaştırma sonucunda güneş takip sistemine ilişkin ışınım ölçer cihazının havanın bulutlu olması durumunda bile güneşin konumunu doğru tespit ettiği ve ölçüm alabildiği, manuel olarak kullanılan ışınım ölçer cihazının ise insani hatalara açık olduğu sonucuna varılmıştır. Güneş takip sistemi üzerine yerleştirilen pirheliyometre cihazına ilişkin ışınım ölçümü ortalama hata yüzde değerlerinin %10’dan daha küçük hatalar ile gerçekleştiği görülmüştür.
Identification of sunflower seeds with deep convolutional neural networks
(Springer, 2020-10-13) Kurtulmuş, Ferhat; Bursa Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; DBP-8176-2022; 15848202900
In the food and agricultural industries, it is crucial to identify and to choose correct sunflower seeds that meet specific requirements. Deep learning and computer vision methods can help identify sunflower seeds. In this study, a computer vision system was proposed, trained, and tested to identify four varieties of sunflower seeds using deep learning methodology and a regular color camera. Image acquisition was carried out under controlled illumination conditions. An image segmentation procedure was employed to reduce the workload in obtaining training images required for training deep convolutional neural network models. Three deep learning architectures, namely AlexNet, GoogleNet, and ResNet, were investigated for identifying sunflower seeds in this study. Different solver types were also evaluated to determine the best deep learning model in terms of both accuracy and training time. About 4800 sunflower seeds were inspected individually for training and testing. The highest classification accuracy (95%) was succeeded with the GoogleNet algorithm.
Immature peach detection in colour images acquired in natural illumination conditions using statistical classifiers and neural network
(Springer, 2014-02) Lee, Won Suk; Kurtulmuş, Ferhat; Vardar, Ali; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; 0000-0001-6349-9687; R-8053-2016; AAH-5008-2021; 15848202900; 15049958800
Detection of immature peach fruits would help growers to create yield maps which are very useful tools for adjusting management practices during the fruit maturing stages. Machine vision algorithms were developed to detect and count immature peach fruit in natural canopies using colour images. This study was the first effort to detect immature peach fruit in natural environment to the authors' knowledge. Captured images had various illumination conditions due to both direct sunlight and diffusive light conditions that make the fruit detection task more difficult. A training set and a validation set were used to develop and to test the algorithms. Different image scanning methods including finding potential fruit regions were developed and used to parse fruit objects in the natural canopy image. Circular Gabor texture analysis and 'eigenfruit' approach (inspired by the 'eigenface' face detection and recognition method) were used for feature extraction. Statistical classifiers, a neural network and a support vector machine classifier were built and used for detecting peach fruit. A blob analysis was performed to merge multiple detections for the same peach fruit. Performance of the classifiers and image scanning methods were introduced and evaluated. Using the proposed algorithms, 84.6, 77.9 and 71.2 % of the actual fruits were successfully detected using three different image scanning methods for the validation set.
Local indications of climate changes in Turkey: Bursa as a case example
(Springer, 2011-04) Vardar, Ali; Kurtulmuş, Ferhat; Darga, Ahmet; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; 0000-0001-6349-9687; AAH-5008-2021; R-8053-2016; 15049958800; 15848202900; 37161164800
This study aims to put out on what ratio Bursa province, one of the important heavy industry regions of Turkey, has been affected climatic process called "Global Warming" or "Climate Change". For this intend climatic measurement results from Bursa center, top of Uludag Mount, YeniAYehir and Keles meteorological stations were used. These measurements were taken as minimum temperature at night-time, maximum temperature at day-time, and mean temperature, mean pressure, insolation intensity, insolation duration, mean wind speed, minimum temperature above soil, soil temperatures at depths of 5, 10, and 20 cm rainfall. Overall, our statistical results showed that there was a considerable warming at statistically 1% and 5% levels in summer months, particularly in July Almost all performed measurements confirm this result. According to climatic data for thirty years (1975-2005), in the last twelve years contrary to previous 18 years, mean temperature values were higher than long-term mean value nine times (years) repetitively. Temperatures did not deviated higher than 0.5A degrees C in six of these. At the temperatures below mean, The maximum deviation was -0.4A degrees C.
Modeling studies on the relation between wind speed and height: Tekirdag sample
(Sıla Science, 2011-07) Eker, Bülent; Durgut, Mehmet Recai; Okur, Ersen; Vardar, Ali; Kurtulmuş, Ferhat; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği.; R-8053-2016; AAH-5008-2021; 15049958800; 15848202900
The experiments of this study were conducted in Degirmenalti Campus of Namik Kemal University in Tekirdag, a province in northwestern Turkey between 2005-2008. An anemograph system which can measure wind speed and wind direction at 2 m and 10 m heights was used in this study. The recorded data was transferred to computer environment with a data logger. The relation between wind speed and height was modeled with the gathered data and it was compared with logarithmic wind speed. Also, artificial neural networks (ANN) method was used in wind speed estimation and error comparisons of the results of measurements and ANN estimations was done. The best results of the error analysis were gained from the developed mathematical model. Especially, the error rate of developed mathematical model for the wind speed values over 2 ms-1 remained under 5%. While error rates appeared to be higher in logarithmic wind speed model, they were quite near to the developed mathematical model in ANN method.
Olgunlaşmamış şeftali meyvesini doğal bahçe koşullarında alınmış görüntülerde görüntü işleme teknikleri ve yapay sınıflandırıcılarla saptayarak sayan algoritmaların geliştirilmesi.
(Uludağ Üniversitesi, 2012) Kurtulmuş, Ferhat; Vardar, Ali; Uludağ Üniversitesi/Fen Bilimleri Enstitüsü/Tarım Makineleri Anabilim Dalı.
Bu çalışmanın amacı ülkemiz için ekonomik değeri yüksek olan şeftali meyvesinin verim haritalamasına yönelik olarak meyvenin erken gelişme döneminde ve doğal ortamından alınmış sıradan renkli görüntülerinden meyveleri tespit ederek sayabilecek algoritmaların geliştirilmesi ve en iyi algoritma performanslarının ortaya koyulmasıdır. Algoritmaların geliştirilmesi ve test edilmesi için görüntüler Bursa Barakfaki köyünde yerel bir çiftçiye ait Elegance Lady çeşidi şeftali bahçesinden alınmıştır. Çalışmada histogram eşitleme ve logaritma dönüşümü gibi görüntü işleme tekniklerinden yararlanılarak doğal koşullarda alınmış görüntülerin aydınlanma koşulları zenginleştirilmiştir. Görüntü işleme tekniklerinden yararlanarak geliştirilen algoritmalar renk, şekil ve doku bilgisini kullanılan öznitelik çıkarma yöntemleriyle görüntülerden çıkarmışlardır. Bu çalışmada kullanılan öznitelik çıkarma yöntemleri, olgunlaşmamış şeftali bitkisini renkli görüntülerde saptama anlamında yenidirler. Çıkarılan özniteliklerle farklı sınıflandırıcıların performanslarını ortaya koymak amacıyla 7 adet sınıflandırıcı eğitilerek denenmiştir. Diskriminant analizi, K-en-yakın komşu, naive Bayes, regresyon ağaçları, sınıflandırma ağaçları, yapay sinir ağları ve destek vektör makinası bu çalışmada kullanılan sınıflandırıcılardır. Görüntülerde arka plan elemesi yapmak ve potansiyel meyve bölgelerini saptamak amacıyla üç farklı görüntü tarama yöntemi geliştirilmiştir. Algoritmaların meyve olarak sınıflandırdığı alt-pencereler blob analiziyle tekilleştirilip meyve sayıları tespit edilmiştir. Farklı meyve tarama yöntemleri, istatistiksel ve deneysel yollarla belirlenen farklı öznitelik birleşimleri, farklı yapay sınıflandırıcılarının kullanımıyla değişik algoritmalar türetilmiş, eğitim ve test setleri üzerinde denemeler gerçekleştirilmiştir. Geliştirilen algoritmaların performansları farklı aydınlanma koşullarını içerecek şekilde karşılaştırılmıştır. Çalışma kapsamında geliştirilen algoritmaların bazılarında % 85'ler düzeyinde saptama başarısı elde edilmiştir. Geliştirilen algoritmalar doğal bahçe koşullarında alınmış görüntülerdeki aydınlanma değişimlerinden fazla etkilenmemişlerdir.
Prediction of lethality by nonlinear artificial neural network modeling
(Wiley, 2016-06-28) Güldaş, Metin; Kurtulmuş, Ferhat; Gürbüz, Ozan; Uludağ Üniversitesi/Karacabey Meslek Yüksekokulu/Gıda İşleme Bölümü.; Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.; Uludağ Üniversitesi/Tıp Fakültesi/Ziraat Fakültesi/Gıda Mühendisliği Bölümü.; 0000-0002-5187-9380; 0000-0001-7871-1628; U-1332-2019; R-8053-2016; K-1499-2019; 35617778500; 15848202900; 8528582100
In this research, the aim was to predict F value (lethality or sterilization value) of canned peas by using a nonlinear auto-regressive artificial neural network model with exogenous input (NARX-ANN). During the model testing, training, validation and reliability steps were followed, respectively. It was found that the model tested was a useful tool to predict the F value for the canned foods with high reliability. Cross-validation rules were performed for training and testing of the model. F value of the 5 kg canned peas could be predicted with a high degree of accuracy (R-2=0.9982, mean square error (MSE)=0.1088) using training the data yielded from 0.5 kg canned peas despite huge mass differences between cross-validated data sets. When the same data sets were trained and tested inversely, a high degree of prediction accuracy (R-2=0.9914, MSE=0.6262) was also observed. The model is also significant in terms of reducing the operational costs due to the fact that higher temperatures and longer process times lead to increased energy costs. Practical ApplicationsIn this research, it was found that nonlinear auto-regressive artificial neural network model with exogenous input is a reliable model for the prediction of lethality rate (F value) in canned food factories. It also provides the advantage of estimating process time more accurately in the retort and thus, reducing operational costs.
Sürekli ve kesikli mikrodalga yöntemleriyle kurutulan elmanın renk değişim analizi
(Bursa Uludağ Üniversitesi, 2020-02-06) Polat, Ahmet; Kurtulmuş, Ferhat; İzli, Nazmi; Bursa Uludağ Üniversitesi/Ziraat Fakültesi/Biyosistem Mühendisliği Bölümü.
Bu çalışmada, sürekli ve kesikli mikrodalga ile kurutulan elma örneklerinin renk değişimleri görüntü işleme yöntemi kullanılarak incelenmiştir. Elma örneklerinden kurutma öncesi ve sonrası olmak üzere toplam yüz adet görüntü alınmıştır. Farklı yöntemlerle kurutulan elma örnekleri RGB renk uzayından L*a*b* renk uzayına çevrildikten sonra analiz edilmiştir. Ayrıca örneklere bir halka maskeleme işlemi uygulanmıştır. Halka maskeli ve halka maskesiz yöntem kullanılarak iki farklı veri elde edilmiştir. Toplam renk değişimi (ΔE), kroma (C), hue açısı (α°) ve kahverengilik indeks (BI) değerleri renk için kinetik parametreleri olarak tanımlanmıştır. Sonuçlar incelendiğinde, kurutmanın tüm örneklerin renk parametreleri üzerine etkisi olduğu görülmüştür. Halka maskeleme yöntemindeki L*, a* ve b* değerlerin halka maskesiz yönteme göre farklı olduğu tespit edilmiştir. Kahverengilik indeksi sonuçları incelendiğinde ise, 200W-KO:3 uygulamasındaki sonucun taze elma ürününe en yakın değer olduğu tespit edilmiştir. Sonuç olarak, sürekli ve kesikli mikrodalga yöntemleri ile elmanın kurutulmasında ürünün renk parametreleri üzerine önemli bir etkisi olduğu ve diğer kurutma yöntemlerine alternatif olabileceği gözlemlenmiştir