Grapevine Image Classification

Keywords: Grapevine Image

Background

For grapevine image classification, a CNN model is constructed, consisting of convolutional layers, pooling layers, and fully connected layers. The convolutional layers are responsible for extracting relevant features from the grapevine images, capturing specific spatial patterns and textures unique to grapevines. Subsequently, pooling layers are employed to reduce the dimensions of these features, enhancing computational efficiency while retaining important information. The fully connected layers then combine the extracted features to perform classification. During training, the model optimizes its parameters using the Adam optimizer and minimizes the categorical cross-entropy loss function. This loss function measures the disparity between the predicted and actual class labels assigned to each grapevine image. To prevent overfitting and improve generalization, techniques like dropout and batch normalization are incorporated. The model's performance is evaluated using the accuracy metric, which assesses its ability to correctly classify grapevine images. The final accuracy achieved by the model is reported as 35 percent. The dataset consists of only 500 images, out of which each 100 set belongs to different classifiers of grapevine. Due to such a shortage of data, the accuracy of the prediction is too low. This may be solved with extra data collection and increasing the training data.

Aim & Objectives

Aim
To Classify the Grapevine Image Using DL Techniques
Objective
Utilize an ANN to accurately classify grapevine images, aiding in disease detection and vineyard management.
Investigate ANN architectures to enhance grapevine image classification, accommodating variations in lighting and angles.
Assess the ANN model's versatility across different grapevine varieties and growth stages for practical vineyard implementation.

Research Methodology

ANN Technique used

Software & Tools

Jupyter, Google Collab

— Preview truncated. The complete document (full chapters, references, diagrams, and appendices) is shared with clients as part of project delivery. —