A Methodological Proposal for Collecting and Creating Macroscopic Photograph Collections of Tropical Woods with Potential for Use in Deep Learning

Abstract

Costa Rica is one of the countries with highest species biodiversity density in the world. More than 2,000 tree species have already been identified, many of which are used in the building, furniture, and packaging industries (Grayum et al. 2003). This rich diversity makes the correct identification of tree species very difficult. As a result, it is common to see in the national market that species are commercialized with mistaken identifications, which makes quality control particularly challenging. In addition, because 90 timber tree species have been classified as "threatened" in Costa Rica, correct identifications are indispensable for law-enforcement.

Using Deep Learning to Identify Costa Rican Native Tree Species From Wood Cut Images

“…To collect and process the wood samples, as well as to create the image database, we developed new protocols that standardize the involved processes. The following two subsections briefly describe the protocols created and first mentioned in Mata-Montero et al (2018).…”

“…Building up the database. Finally, metadata such as date and place of extraction, specimen ID, dimensions, and weight, among others, are recorded for each cube (Mata-Montero et al, 2018). Additionally, for each picture, the picture itself, the type of cut (cross, tangential or radial), cube it belongs to, and stage of the wood when the picture was taken (fresh, dry) is recorded and stored.…”