Assessment of soil properties using microscope based computer vision
Bibliographic record
Abstract
ABSTRACTSoil texture and organic matter content are important indicators of the quality and health of soil. They affect a range of soil properties and processes, which are fundamental for agriculture and civil engineering. Several traditional methods and advanced measurement techniques aim to address the challenge of quantifying these attributes. However, their cost, time requirements, sophisticated analytical methods and in-situ inapplicability pose a major challenge to rapid measurement. This research discloses the development of a new, inexpensive, microscope-based sensor system to estimate content of both sand and organic matter. The research was divided into two experiments conducted over a span of two years, each approaching the problem from two different computational perspectives. The first experiment involved images of air dried soil samples from Field 26 (acquired in 2014, organic soil) of the Macdonald Campus Farm of McGill University. The set of images was analyzed for sand and organic matter content using color and spatial image analysis, then validated against data obtained using conventional methods in a laboratory. Predictive relationships were developed using simple linear regressions based on parameters computed from the acquired imagery, such as hue, saturation, value, porosity, and variance estimates. The best sand and organic matter prediction models exhibited coefficients of determination (R2) values of 0.63 and 0.83, respectively, with RMSE = 84.7 g/kg for sand content and 0.11 for log SOM. In addition to the first set of images, the second experiment explored both laboratory and in situ measurements from Field 86 (acquired in 2015, mineral soil). This method used a continuous wavelet transform to characterize sand content which was in strong agreement with the laboratory measurements (r2 = 0.86 and RMSE = 44.7 g/kg for organic soil; r2 = 0.87 and RMSE = 40.2 g/kg for mineral soil). However, the efficiency of this algorithm was subpar for the images collected in-situ (r2 = 0.48 and RMSE = 80.6 g/kg). This was due to the excessive soil water content, which can be addressed by modifying the microscope holder design and data collection protocol. The portable nature of the image acquisition system and the good performance of the wavelet algorithm shows promise for the future use of the system to rapidly quantify key soil physical attributes.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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; a candidate call from one teacher head, not a consensus.
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".