Detection and Recognition of Pollen Grains in Multilabel Microscopic Images

Kubera, Elżbieta; Kubik-Komar, Agnieszka; Kurasiński, Paweł; Piotrowska-Weryszko, Krystyna; Skrzypiec, Magdalena

doi:10.3390/s22072690

Cited by 24 publications

(11 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One is YOLOv8m, with the highest mAP50 value of 96.7%, recall value of 94%, and inference speed of 5.7 ms, while the other option is YOLOv8s, with the mAP50 value of 96.5%, the recall value of 93.6%, and the best inference speed of 2.3 ms. The study's results are very similar to those achieved by Kubera et al [33]. Though they used the older version of YOLO (YOLOv5) to detect three classes of pollen grains, YOLOv5l (large) performed the best, with an mAP50-95 value of 91.5%.…”

Section: Discussionsupporting

confidence: 85%

Application of YOLOv8 and Detectron2 for Bullet Hole Detection and Score Calculation from Shooting Cards

Butt,

Glas,

Monsuur

et al. 2023

View full text Add to dashboard Cite

Scoring targets in shooting sports is a crucial and time-consuming task that relies on manually counting bullet holes. This paper introduces an automatic score detection model using object detection techniques. The study contributes to the field of computer vision by comparing the performance of seven models (belonging to two different architectural setups) and by making the dataset publicly available. Another value-added aspect is the inclusion of three variants of the object detection model, YOLOv8, recently released in 2023 (at the time of writing). Five of the used models are single-shot detectors, while two belong to the two-shot detectors category. The dataset was manually captured from the shooting range and expanded by generating more versatile data using Python code. Before the dataset was trained to develop models, it was resized (640 × 640) and augmented using Roboflow API. The trained models were then assessed on the test dataset, and their performance was compared using matrices like mAP50, mAP50-90, precision, and recall. The results showed that YOLOv8 models can detect multiple objects with good confidence scores. Among these models, YOLOv8m performed the best, with the highest mAP50 value of 96.7%, followed by the performance of YOLOv8s with the mAP50 value of 96.5%. It is suggested that if the system is to be implemented in a real-time environment, YOLOv8s is a better choice since it took significantly less inference time (2.3 ms) than YOLOv8m (5.7 ms) and yet generated a competitive mAP50 of 96.5%.

show abstract

Section: Discussionsupporting

confidence: 85%

Application of YOLOv8 and Detectron2 for Bullet Hole Detection and Score Calculation from Shooting Cards

Butt,

Glas,

Monsuur

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Indeed, similar tasks are possible in other biological image contexts [e.g. multilabel segmentation ( Kubera et al , 2022 ), or dealing with overlapping via instance segmentation ( Saleh et al , 2019 )]. Other CNN-minirhizotron applications can even tackle this last problem, albeit without the corrective annotation which made our approach useful for automatic data ( Peters et al , 2022 ).…”

Section: Discussionmentioning

confidence: 99%

High frequency root dynamics: sampling and interpretation using replicated robotic minirhizotrons

Nair

Strube

Hertel

et al. 2022

Journal of Experimental Botany

View full text Add to dashboard Cite

Automating dynamic fine root data collection in the field is a longstanding challenge with multiple applications for co-interpretation and synthesis for ecosystem understanding. High frequency root data are only achievable with paired automated sampling and processing. However, automatic minirhizotron (root camera) instruments are still rare and data is often not collected in natural soils nor analysed at high temporal resolution. Instruments must also be affordable for replication and robust under variable natural conditions. Here, we show a system built with off-the-shelf parts which samples at sub-daily resolution. We paired this with a neural network to analyse all images collected. We performed two mesocosm studies and two field trials alongside ancillary data collection (soil CO2 efflux, temperature and moisture content, and ‘PhenoCam’-derived above-ground dynamics). We produce robust and replicated daily time-series of root dynamics under all conditions. Temporal root changes were a stronger driver than absolute biomass on soil CO2 efflux in mesocosm. Proximal sensed aboveground dynamics and belowground dynamics from minirhizotron data were not synchronised. Root properties extracted were sensitive to soil moisture and occasionally to time of day (potentially relating to soil moisture). This may only affect high frequency imagery and should be considered in interpreting such data.

show abstract

“…The next paper [ 11 ], written by E. Kubera, A. Kubik-Komar, P. Kurasiński, K. Piotrowska-Weryszko, and M. Skrzypiec, describes the application of deep neural network YOLO. (You Only Look Once) models for detection and recognition of pollen grains.…”

Section: Overview Of the Contributionsmentioning

confidence: 99%