Skip to main content
Advertisement

< Back to Article

Fig 1.

A digital image of an herbarium specimen.

The plant and original label containing the original metadata information, as well as additional reference information for the digital image: a ruler for scale and a color reference grid.

More »

Fig 1 Expand

Fig 2.

Overview of the confidence-based workflow.

By only considering labels over a certain probability threshold, we increase the final accuracy of the model at the cost of coverage on the overall dataset (red: wrong label, green: true label, gray: rejected label).

More »

Fig 2 Expand

Fig 3.

Rejection/Accuracy curves for all four models.

Associated labels: (a) Budding, (b) Flowering, (c) Fruiting, (d) Reproductive.

More »

Fig 3 Expand

Fig 4.

DoY estimation error and number of empty species estimates as a function of the confidence threshold.

More »

Fig 4 Expand

Fig 5.

Comparison of study ground-truth and AI estimate, per species, with a confidence threshold of 0.5.

Blue: study ground truth with mean/std, red: AI estimate with mean/std.

More »

Fig 5 Expand

Fig 6.

Comparison of study ground-truth and AI estimate, per species, with a confidence threshold of 0.99.

Blue: study ground truth with mean/std, red: AI estimate with mean/std.

More »

Fig 6 Expand

Fig 7.

Rejection/Accuracy curves of the INaturalist2018 multi-class classifier.

More »

Fig 7 Expand

Table 1.

Expended accuracy results on validation set of INaturalist2018 with confidence thresholds.

Accuracy and rejection rate percentages for particular minimum confidence thresholds.

More »

Table 1 Expand

Fig 8.

Workflow to automatically obtain a macrophenological analysis of flowering time shifts: from non-annotated samples to regional trends.

More »

Fig 8 Expand

Fig 9.

Flowering shifts distribution per growth form status.

More »

Fig 9 Expand

Fig 10.

Flowering shifts distribution per native/introduced status.

More »

Fig 10 Expand

Fig 11.

Flowering shifts distribution per wetland status.

More »

Fig 11 Expand

Fig 12.

Flowering shifts distribution per early/late season status.

More »

Fig 12 Expand

Fig 13.

Flowering shifts distribution per season spread status.

More »

Fig 13 Expand

Table 2.

Results of Welch’s T-test evaluating effect of plant characteristics on flowering time shift.

More »

Table 2 Expand

Table 3.

Results of Pagel’s test for correlated evolution between two binary characters.

More »

Table 3 Expand

Fig 14.

Phylogenetic tree with seasonality character and detected flowering shifts.

More »

Fig 14 Expand

Table 4.

Accuracy results on validation set.

More »

Table 4 Expand

Table 5.

Extended accuracy results on validation set with confidence thresholds.

Accuracy and rejection rate percentages for particular minimum confidence thresholds.

More »

Table 5 Expand

Fig 15.

2D projection of the Flowering classifier’s embeddings for the validation dataset from a t-SNE algorithm.

(a) Blue and Green are the Flowering/Not Flowering classes respectively. (b) Additional overlay with the confidence scores.

More »

Fig 15 Expand

Fig 16.

Examples of images with Sure and Unsure Flowering classification.

(a,b) High Flowering Confidence (a: 97%, b: 94%), (c,d) High Not-Flowering Confidence (a: 98%, b: 99%), (e, f) Unsure Flowering (e: 41%, f: 62%).

More »

Fig 16 Expand

Table 6.

Overview of the classes distribution and amount of data in the INaturalist2018 dataset.

More »

Table 6 Expand

Fig 17.

Model annotation estimated trends for each threshold (Black, Brown, Red, Orange, Yellow, Green, Blue), human annotations estimated trend (Purple).

More »

Fig 17 Expand

Fig 18.

Amount of slope error (between the automatically annotated and the human annotated ground-truth) against the number of available samples to perform the regression.

Each dot corresponds to a given species for a given threshold.

More »

Fig 18 Expand

Table 7.

National Wetland Plant List Wetland Indicator Status categories and definitions.

See [78] for more in-depth definitions and other information.

More »

Table 7 Expand