Crop Disease Detection Using Deep Learning Techniques on Images
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Agriculture plays a crucial role in the economic development of many countries and sustains the global population despite facing various challenges like climate change, pollinator decline, and plant diseases. These threats to food security highlight the need for innovative solutions to prevent crop loss. Leveraging smartphone technology for automated image recognition-based disease diagnosis has emerged as a promising approach, thanks to their computing power and high-resolution cameras. To address this issue, we have focused on deep learning-based image detection techniques to identify plant diseases using the "PlantVillage" dataset. Several deep learning architectures, including AlexNet, GoogleNet, ResNet50, and InceptionV3, were employed and trained using two approaches: 'Training from scratch' and 'transfer learning’. The results of the analysis reveal GoogLeNet architecture achieved the highest accuracy of 0.999 for color images and 0.996 for segmented images, whereas InceptionV3 trained from scratch gave the highest accuracy of 0.994 for grayscale images with a train-test ratio of 90:10. All the models trained from scratch achieved the maximum F1-score of 1.0 for color and segmented images whereas for grayscale images, GoogleNet and InceptionV3 achieved the highest F1-score of 0.999 with train-test ratio 90:10. These findings indicate the potential of deep learning methods in detecting and diagnosing plant diseases, which can significantly enhance the efficiency and accuracy of disease diagnosis processes in agriculture. Further research and improvements in image recognition techniques can lead to more robust and effective solutions for securing global food production.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it