Body condition score (BCS) has been a useful tool in estimating the health of cattle for many years now. This categorical metric requires experienced observers to visually inspect cows and assess body fat deposits regularly via a time consuming, subjective process. Low cost RGB+depth cameras have been used alongside machine learning algorithms in the past and have shown great promise, however, more advanced techniques are projected to yield better performance. In this work, a vision transformer (ViT) is pretrained using a recently developed self-supervised pretraining method, masked image modeling, and then fine-tuned on RGB+depth BCS data with the objective of improving performance. Model accuracy was found to be highly dependent on dataset curation, ranging from 64% to 92% accuracy. These discrepancies are attributed to non-unique data in the training and test splits and an inherently unbalanced dataset, both of which are discussed in detail. It is recommended that engineers and animal scientists collaborate more closely, as certain details related to dataset curation are critical to thoroughly assess performance and robustness of automated methods for BCS determination.
Accurate estimation of one-seed juniper (Juniperus monosperma [Engelm.] Sarg.) intake by herbivores often requires harvesting, transporting, and storing plant material that is later used in pen experiments. Such manipulation could alter terpenoid profiles and modify herbivory levels significantly. We used gas chromatography mass spectrometry (GC/MS) to analyze the terpenoid profile of leaves from 10 short (0.5 m +/- 0.05, mean +/- SE) and 10 tall (1.14 m +/- 0.06) one-seed juniper saplings subjected to 3 handling protocols: a) placed on dry ice after clipping and stored after 5 hours at –80 degreesC for 3 weeks (Control); b) kept at ambient temperature for the first 24 hours and then frozen at –80 degreesC for 3 weeks; or c) kept at ambient temperature for the first 24 hours, and then stored at 8 degreesC for 3 weeks. Juniper saplings contained 51 terpenoids, 3 of which were unknown compounds. Fourteen terpenoids accounted for 95% of the total amount of volatiles. The most abundant compound was a-pinene, which accounted for 65% of total terpenoids present. Handling protocols were not associated with detectable differences in total terpenoid content (Means +/- SE, Control: 21.68 +/- 1.42 mg g-1 dry matter [DM]; Frozen after 24 hours: 19.55 +/- 1.08 mg g-1 DM; Refrigerated after 24 hours: 18.80 +/- 1.13 mg g-1 DM). However, total terpenoid amount and concentration of a few major compounds tended to decrease with increasing storage temperature. Handling protocols induced detectable variations in a small number of minor terpenoids. We observed large among-plant variation in terpenoid profiles that was not fully explained on the basis of sapling size. This study suggests that the length of storage period of one-seed juniper branches should not exceed 3 weeks and that storage refrigeration temperatures should be kept below 8 degreesC to prevent significant alterations in terpenoid profiles.
Photoluminescence of plant extract solutions has been investigated for discrimination of plant life forms (grasses, forbs, and shrubs) using principal component analysis (PCA). Clippings from each of six plant species representing three different plant life forms potentially found in the diets of free-ranging herbivores in the Chihuahuan Desert of North America were investigated for possible discernment. These plants included Sporobolus flexuosus (mesa dropseed, a grass), Pleuraphis mutica (tobosa, a grass), Dimorphocarpa wislizenii (spectacle pod, a forb), Sphaeralcea incana (pale globemallow, a forb), Flourensia cernua leaves (tarbush, a shrub), and Atriplex canescens leaves and stems (fourwing saltbush, a shrub). Emission spectra (370–600 nm) from phosphate buffered saline (PBS) extract solutions (pH 2.2, 7.5 and 12.5) were recorded for each plant with excitation at 365 nm. Use of PBS minimized chlorophyll interference. Discernment among plant life forms within these plant species was achieved.