Sunday, November 27, 2022
spot_img
HomeEducationImprove Understory and Overstory Tree Height Reliability With Agisoft Metashape

Improve Understory and Overstory Tree Height Reliability With Agisoft Metashape

agisoft metashape depth filtering. Using the latest technology in 3D image analysis, Metashape improves the accuracy of understory and overstory tree height measurements. The software works by aligning photos and creating a sparse point cloud. However, this method also compresses the data by a factor of 100, leading to a loss of resolution. As such, it is important to weigh the computational time against the output quality before making the decision.

Improved understory tree height reliability

Agisoft Metashape’s depth filtering feature sets can improve understory tree height reliability in a forest. Our studies have shown that a High-Quality depth map filter setting can identify up to 84.4% of the understory trees in a forest. However, this accuracy is reduced when depth map filtering is performed with an Aggressive filter setting. These results suggest that other depth filtering feature sets may improve the accuracy of the overstory trees.

Detection reliability of understory trees increased along the gradient from coarse to fine image resolutions. The mean underestimation bias increased from -0.03 to -0.19 m, but not significantly. The F-scores for detected understory trees increased from 33% to 24% with a shift from Lowest to Ultra High Quality. These improvements are not statistically significant, but they support the recommendations to retain the original image resolution for optimum results.

The same depth map feature sets have been shown to improve the visual representation of understory trees. This is especially true of smaller trees. However, these feature sets can produce noise points because of moving vegetation and trees. Further work is needed to understand whether this effect can be translated to other complex forest systems.

Using Metashape’s depth map feature sets to improve understory tree height reliability is a good way to minimize noise. Using this feature set, users can reduce the number of point observations and increase the accuracy of the model. In addition, users can customize the settings to remove outlier point observations. This feature allows them to use depth map feature sets and segmented images based on their distance from the camera. agisoft metashape depth filtering.

Improved detected overstory tree height reliability

The Agisoft Metashape depth map filter has a limited impact on detected overstory tree height reliability. However, it did decrease performance significantly at the Aggressive depth filter setting. This may be due to the fact that the depth map filter uses connected component size, which does not accurately represent the irregularity of tree crown structures.

We also observed that the detection performance of overstory trees improved along a gradient from coarsest to finest image resolution. Moreover, the F-score for detected overstory trees improved from 0.172 to 0.717. These results were similar when we applied Medium, High, and Ultra High-Quality settings.

Improved detected overstory tree height reliability with Agisoft Metashape depth filtering has several advantages. First, it improves the structural variability of the depth map, while increasing the amount of image overlap improves the quality of the point cloud. Second, it decreases the level of noise in the image and therefore improves the visual representation of vegetation. Finally, this method reduces tinning errors.

A further improvement is the detection of false positives and omissions. The density of the dense cloud is an important determinant of false positives and omissions in understory tree height. density metrics of the cloud are also useful for evaluating the sensitivity of the method. The density metrics are analyzed based on two-way ANOVA with Tukey’s HSD test.

The results of this research provide important information for managers to decide whether they should implement this method on a large scale. For example, it can be used for monitoring the growth of ponderosa pine forests. Forested environments may be particularly challenging to assess. The depth filtering suggestions are not well documented. In such a scenario, the proposed method may be effective for monitoring ITD.

RELATED ARTICLES

LEAVE A REPLY

Please enter your comment!
Please enter your name here

- Advertisment -

Most Popular

Recent Comments