Smart Computing and Systems Engineering (SCSE)

Permanent URI for this communityhttp://repository.kln.ac.lk/handle/123456789/18936

Browse

Subject: search.filters.subject.Image classification

Search Results

Now showing 1 - 3 of 3
  • No Thumbnail Available
    Item
    Vehicle type validation for highway entrances using convolutional neural networks
    (International Research Conference on Smart Computing and Systems Engineering - SCSE 2018, 2018) Juwanwadu, L.N.W.; Jayasiri, A.
    Vehicle type validation for Highway entrances using convolutional neural networks is an approach taken to automate the highway toll systems of Sri Lanka. Available automated highway toll systems in the world use sensor-based validation systems to validate the vehicles that are entering the highways. Mainteneance cost of these systems is high. A vision-based validation system has not been implemented, as yet. This paper introduces a vision-based method to validate vehicles for highway systems which can reduce the cost while increasing the efficiency and safety. A Convolutional Neural Network (CNN) model was developed to achieve this objective. The CNN model employed here uses a binary classification to categorize vehicles as allowed vehicles and non-allowed vehicles for entering the highway. The model developed here showed 86.69% accuracy. The model was manually tested for different vehicle types using a GUI based application and all the test images were successfully classified into their classes.
  • No Thumbnail Available
    Item
    Gender recognition of Luffa flowers using machine learning
    (International Research Conference on Smart Computing and Systems Engineering - SCSE 2018, 2018) Gunasinghe, H.N.; de Silva, R.
    Automatic flower gender identification could be introduced to large farmlands to help artificial pollination of imperfect flowers. Incomplete flowers contain either male or female organs but not both. In this paper, we present a computer aided system based on image processing and machine learning to identify the gender of a Luffa flower automatically. A pre-trained machine learning model is used for gender segmentation of flowers. The system is developed using Tensorflow Machine Learning Tool, which is an open-source software library for Machine Intelligence. The network was selected as the Google’s Inception model and a dataset was prepared after capturing flower images from a Sri Lankan Luffa farm. The system was tested using two datasets. The first contained the captured original images and the second was prepared by cropping each image to extract male and female floral organs, stamen and pistil respectively. The prototype system classified the flowers as either male or female at 95% accuracy level. The experimental results indicate that the proposed approach can significantly support an accurate identification of the gender of a Luffa flower with some computational effort.
  • No Thumbnail Available
    Item
    Identification of water stressed leaves using Artificial Intelligence: The case of eggplant
    (International Research Conference on Smart Computing and Systems Engineering - SCSE 2018, 2018) Senanayake, P.A.; de Silva, R.
    Identification of water stress of leaves from the photos taken has a long history. Researchers have defined a parameter called Leaf Water Content (LWC) to quantify the dryness of leaves. However, in the case of automatic watering of plants, such high accuracy of LWC is not needed as a decision to water or not alone is sufficient. Furthermore, the agricultural industry cannot use methods of remote sensing that are required to find LWC as they are complex and costly. In the current practice, farmers use their knowledge and experience together with the appearance of plants to estimate the water stress and watering time point of plants. The approach presented in this paper is easily implemented and requires only a series of photos taken by a smartphone or a camera and a software app. In this paper, a method s introduced using Artificial Intelligence (AI) where the images of leaves are directly used to determine whether the leaves are water stressed. We could identify the water stressed leaves accurately using this method. Once an app based on our method is developed, it could easily be used by farmers to automatically identify whether the eggplants are water stressed and need watering.