Full text
Title Plant Seedlings Classification Model Abstract This study presents a Convolutional Neural Network (CNN)-based deep learning approach to automate the classification of plant seedlings from image data. Using the Aarhus Plant Seedlings Dataset, the model identifies 12 plant species to assist in early-stage crop monitoring and weed management. The system significantly reduces manual effort, enhances classification accuracy, and supports sustainable agricultural practices by enabling scalable, data-driven field analysis. Keywords Deep Learning, Computer Vision, CNN, Agriculture, Seedling Classification, TensorFlow, Keras, Aarhus Dataset Model & Publication Details Attribu te Description Model Plant Seedlings Classification — AI for Smarter Agriculture Author Joyjit Roy ([email protected])
Huggin Face Link https://huggingface.co/joyjitroy/Plant_Seedlings_Classification GitHub Link https://github.com/joyjitroy/Machine_Learning/blob/main/Plant_Seeding_Classifica tion_using_CNN.ipynb Publish ed Date July 2024 Problem Context Identifying and categorizing plant seedlings manually remains a time-intensive and error-prone task in agriculture. Despite technological advancements, farmers still rely heavily on visual inspection to monitor crops and detect weeds. This manual dependency limits scalability, delays decision-making, and affects yield prediction accuracy. There is a pressing need for automated, consistent, and scalable methods to improve early-stage crop monitoring and agricultural efficiency. Solution Overview This project builds a Convolutional Neural Network (CNN) to classify plant seedlings into their respective categories, reducing manual effort and enabling scalable, automated monitoring in agriculture. Model Attributes Attribute Description Model Type CNN-based Image Classifier Model File Plant_Seeding_Classification_using_CNN.zip Framework TensorFlow / Keras Dataset Aarhus Plant Seedlings Dataset (12 plant species) Data Files images.npy (image array) Labels.csv (species labels) Input RGB image of a seedling (128 × 128) Output Predicted plant species label Data Dictionary Feature/File Description images.npy A NumPy binary file that stores arrays efficiently. It preserves the shape, datatype, and structure of image data, allowing fast loading and processing during model training and inference Labels.csv CSV file containing labels for each image
Label Column specifying plant species name Image Shape 128 × 128 × 3 RGB Classes 12 species List of Species Black-grass, Charlock, Cleavers, Common Chickweed, Common Wheat, Fat Hen, Loose Silky-bent, Maize, Scentless Mayweed, Shepherds Purse, Small-flowered Cranesbill, Sugar beet Evaluation Metrics Metric Description Accuracy, Precision (weighted), Recall (weighted), F1-Score (weighted) Evaluate overall classification performance across all 12 plant species. Confusion Matrix Visualizes class-wise prediction performance and misclassifications . Cross-Entropy Loss Used as the model’s training objective to optimize prediction confidence . Training/Validation Accuracy Curve Tracks convergence and overfitting trends during model training . Hyperparameter Tuning GridSearchCV and manual tuning performed with f1_weighted scoring to select optimal architecture and learning parameters . Why It Matters Reduces manual effort in plant identification Boosts accuracy and consistency in seedling detection Supports sustainable, data-driven agriculture Enables automation in large-scale farming and greenhouse monitoring Glossary Term Definition CNN (Convolutional Neural Network) A type of deep learning algorithm particularly effective for image recognition and classification tasks. TensorFlow / Keras Open-source frameworks used to build, train, and deploy neural network models efficiently.
NumPy (.npy) A binary file format used to store large multidimensional arrays and numerical data in Python. It retains array shape and datatype, allowing fast load and computation. Aarhus Plant Seedlings Dataset A public dataset containing images of 12 common crop and weed species used for image classification tasks in agricultural research. RGB Image (128×128×3) A color image represented with three color channels (red, green, blue) resized to 128×128 pixels for model input. Label The class name or category assigned to each image sample, indicating the type of seedling. Precision Agriculture A farming management approach that uses AI, sensors, and data analytics to optimize crop yields and resource use. Disclaimer - This document is a public preprint describing the Plant Seedlings Classification Model for open research and reproducibility.
