Comparing the Performance of Different Classifiers for Urban Change Detection: A Case Study in Kingston, Ontario
Bibliographic record
Abstract
Abstract. Remote Sensing-based change detection (CD) focuses on the identification of Earth's surface transformations through analysis of multi-temporal satellite images captured for the same geographic region at different points in time. Two common classification-based change detection techniques are post-classification comparison of land cover maps and direct classification of image differences between two periods of interest. In either approach, the selection of an appropriate classifier is critical. Consequently, this study focuses on assessing the performance of different classifiers, including the multi-layer perceptron neural network (MLPNN), support vector machine (SVM), random forest (RF), maximum likelihood (MLH), k-nearest neighbor (KNN), and gaussian naïve bayes (GNB). For this evaluation, two high resolution images captured in 2016 and 2020 by the PS2 sensor within the PlanetScope satellite constellation were used. Additionally, a novel unbiased sampling technique was introduced to selectively capture a minimal number of reference pixels. In the context of change detection, slight variations were observed in classification performance rankings between post-classification and difference image classification methods. However, a consistent trend emerged. The MLPNN consistently achieved the highest accuracy, closely followed by RF and SVM as the second or third-best performers in each technique. In contrast, GNB consistently yielded less favourable results. Importantly, our findings highlight the persistent superiority of the difference image classification in terms of change detection accuracy across all six classifiers. Furthermore, this method offers a significant advantage due to its reduced processing time and computational demands, positioning it as the preferred choice for binary change detection when compared to post-classification techniques.
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.
How this classification was reachedexpand
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 | 0.003 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.002 | 0.003 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; both teacher heads agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".