http://repository.mut.ac.ke:8080/xmlui/handle/123456789/2882 2026-04-05T10:12:32Z http://repository.mut.ac.ke:8080/xmlui/handle/123456789/6400 A Support Vector Machine and Artificial Neural Network Model for Enhanced Image Classification of Maize Leaf Diseases Ochango, Vincent Mbandu Image classification accuracy is the total number of images predicted correctly out of the total images in the test dataset in the field of computer vision. Classifying the images accurately is still a challenge due to single image classification models being biased and having high variance. The research created a combination of two models (Artificial Neural Network + Support Vector Machine), the maize leaf disease image features that were extracted were passed to the developed model which classified the diseases with high accuracy compared to the single models. Dimensionality reduction was also considered to reduce the computational complexity and this was achieved by using the Histogram of Oriented Gradient feature descriptor which extracted only relevant features and through away information that was not necessary. The relevant features were considered as the key point since an image was differentiated from each other using the key points. The developed model input was the features extracted which were in a form of a vector space known as an array of numbers and each number represented a particular feature. The developed image classification model consists of two modules; the feature extraction module and the image classification module. The feature extraction module was integrated to work together with the classification module and the features extracted by the feature extraction module were normalized to make them scale-invariant and less susceptible to light which is one of the factors that usually affects image classification accuracy. The classification module was also adjusted by combining two classifiers; Artificial Neural Network and Support Vector Machine and the main reason were for the Support Vector Machine to replace the softmax layer used for classification in the Artificial Neural Network since the Support Vector Machine have the hyperplane component which is a line that accurately separates data belonging to different classes and this made SVM to classify maize leaf disease images accurately. The Support Vector Machine also has the capability of minimizing the generalization error on unseen data which resulted in better prediction results. The common rust, leaf spot, and northern leaf blight and healthy images were used during the feature extraction process, training, and validation of the model. The feature extraction methods were compared on how they perform with image classification models to find out which feature descriptor performs best. The experimental results indicated that the Histogram of Oriented Gradients performs well with the image classifiers compared to KAZE and Oriented FAST and Rotated BRIEF and the Histogram of Oriented Gradients method reduces computational complexity during the image feature generation process. The model which was a combination of three methods, Histogram of Oriented Gradient, Artificial Neural Network, and Support Vector Machine emerged the best in terms of image classification. The experimental outcome based on performance metrics indicated that the developed model had a 0.95 accuracy score. The experimental result shows that the Histogram of the Oriented Gradient together with Artificial Neural Network and Support Vector Machine classifier is the best combination model for maize leaf disease identification since it produced the highest accuracy score compared to the other image classification models. The researcher finally recommends the model to be used today and in the future when it comes to classifying maize leaf disease images. Master of Science in Information Technology, 2022 2022-10-01T00:00:00Z http://repository.mut.ac.ke:8080/xmlui/handle/123456789/6387 Enhanced Deep Learning Model to Detect Anomalies in Surveillance Videos Munyua, John G. Increased security challenges and advancements in technology have led to heavy usage of surveillance cameras. This has resulted in an overwhelming abundance of video data which requires automated analytics for better utilization. The big volume of the video data generated by surveillance devices presents an enormous problem to the security personnel since they must monitor the footage frame by frame to identify the abnormal activities (security threats) like violence, and thuggery, among others. Successful identification of anomalies in surveillance footage will ease the work of Closed-Circuit Television (CCTV) operators greatly since they can search through a big volume of the video data easily. Another importance of this research is the contribution to computer vision since the model can be applied in other areas like robotic surveillance or unmanned surveillance. There have been attempts to automate the surveillance process using smart surveillance. However, these solutions are challenged due to high error rates and inefficiency while identifying abnormal scenes. Modern automated video analytics, use deep learning algorithms like; Convolutional Neural Networks (CNN), Long-Short Term Memory (LSTM), convolutional LSTM and 3DCNN.These approaches have their strengths and weaknesses, and it becomes a research challenge to determine the best model to use in detecting anomalies. Another challenge presented herein is the accuracy of detecting anomalies in surveillance videos. A comparative study was carried out to cross-examine deep learning models used in anomaly detection. Empirical data was collected to measure the accuracy of the deep learning models in anomaly detection. The best model was determined by analyzing the accuracies of the model published since 2016. Experiments were set up in Google Collab and Google Cloud. These environments were configured to use Python 3.7, Keras and TensorFlow machine learning frameworks. The study improved the selected deep learning model through, optimization of the model structure and depth tuning. The study found that deeper autoencoders have high prediction accuracy and deeper spatial autoencoders draws more features from the videos and that increases their accuracy. Validation of the enhanced model was done through further experiments that compared the prediction accuracy acquired from the enhanced model against the existing model set as the control group. Their Receiver Operating Characteristic Curve (ROC) scores from UCSD Ped1 and Ped2 datasets were compared. Comparative analysis of the recorded model accuracies was tabulated and a percentage increase in the model accuracy was noted. A sign test was used to test the significance of the improvement and at both 1% and 5% significance levels, empirical evidence of the enhancement was found. This work contributed to the autoencoder design paradigms, improvement of Spatial-Temporal Autoencoder accuracy through depth and regularization tuning and reduction of anomaly detection errors in surveillance videos. The study has shown that the depth of spatial-temporal autoencoder impacts its anomaly prediction accuracy. In future work, integration of continual learning and real-time anomaly detection should be considered. Master of Science in Information Technology, 2022 2022-08-01T00:00:00Z http://repository.mut.ac.ke:8080/xmlui/handle/123456789/6379 A Transfer Learning and Two-Level Hyperparameter Optimization Based Model for Improved Classification of Diabetic Retinopathy Wambugu, Jackson K. Automated diagnosis of disease from medical images using machine learning has been in rise in the recent past. One such case is the classification of diabetic retinopathy from fundus images. Diabetic Retinopathy is an eye disease that is a result of diabetes mellitus and it is major cause of blindness among people of the working age. Diabetic retinopathy has five main classes namely: No DR, Mild DR, Moderate DR, Severe DR, and Proliferative DR. Deep learning has been used previously in this field and it has proved to be better than conventional machine learning approaches. However, deep learning involves training a model from scratch thus making it to be data hungry, require high training cost, have poor generalizability, and they don’t deliver high performance. Meta-learning also known as learning-to-learn is a field of machine learning which aims at improving deep learning by enabling models to improve their performance capabilities and reduce training cost. Meta-learning techniques include multi-task learning, transfer learning, self-optimization, and few-shot learning. Several transfer learning architectures pre-trained on the ImageNet dataset have been used by different researchers and they have demonstrated superior performance over deep learning. However, domain-shift generalizability and optimal performance of pre-trained architectures are major challenges facing transfer learning. This so because these models are not properly tuned for cross-domain optimality. The aim of this study was to develop an improved model for classification of diabetic retinopathy into its five classes. To achieve this, the researcher used the following approach: A VGG16 network pre-trained in ImageNet was modified such that the top-layer was rebuilt and an attention model was added. Two-level optimization was used during training in which the model was allowed to self-tune its learning rate based on the training parameters. The EyePACS dataset obtained from Kaggle repository was used in training, validating, and testing the model. The model was developed in Google Collaboratory platform using python programming language, TensorFlow, and Keras. The study achieved the following results: Accuracy 89.06%, Precision 88.9%, Recall 89.2%, F1-Score 75%, Quadratic Cohen Kappa Metric 0.84, Area Under the Curve (AUC) 93.3%. The results of the study demonstrated improved performance compared to other existing models in literature such as Qummar et al (2019), Jinfeng et al (2020), Chilukoti et al (2022), that classify diabetic retinopathy into five classes. The study concluded that leveraging on previously acquired knowledge and efficient optimization of neural networks using data driven self-optimization delivers better performance than conventional machine learning and deep learning. In future researchers can consider using reinforcement learning and transfer learning in classification of diabetic retinopathy. Master of Science in Information Technology, 2022. 2022-10-01T00:00:00Z http://repository.mut.ac.ke:8080/xmlui/handle/123456789/3022 Factors Influencing Digital Divide In the Rural-And Semi-Urban Regions of Kenya: Case Study of Juja Town and the Surrounding Areas Ndung’u, George M. Master of Science in ICT Policy and Regulations, 2010 2010-01-01T00:00:00Z