SMACC: A System for Microplastics Automatic Counting and Classification

Abstract

The management of plastic debris is a serious issue due to its durability. Unfortunately, million tons of plastic end up in the sea becoming one of the biggest current environmental problems. One way to monitor the amount of plastic in beaches is to collect samples and visually count and sort the plastic particles present in them. This is a very time-consuming task. In this work, we present a Computer Vision-based system which is able to automatically count and classify microplastic particles (1-5 mm) into five different visual classes. After cleaning a collected sample in the lab, the proposed system makes use of a pair of its images with different characteristics. The procedure includes a segmentation step, which is based on the Sauvola thresholding method, followed by a feature extraction and classification step. Different features and classifiers are evaluated as well as a deep learning approach. The system is tested on 12 different beach samples with a total of 2507 microplastic particles. The particles of each sample were manually counted and sorted by an expert. This data represents the ground truth, which is compared later with the results of the automatic processing proposals to evaluate their accuracy. The difference in the number of particles is 34 (1.4%) and the error in their classification is less than 4% for all types except for the line shapes particles. These results are obtained in less than half of the time needed by the human expert doing the same task manually. This implies that it is possible to process more than twice as many samples using the same time, allowing the biologists to monitor wider areas and more frequently than doing the process manually. INDEX TERMS Computer vision, deep learning, microplastics classification.

A Practical Overview of Methodologies for Sampling and Analysis of Microplastics in Riverine Environments

“…This process requires considerable time and resources in terms of researchers involved in counting hundreds of particles per sample and a high risk of data overestimation for false positives [10,11,67,116,117]. To this purpose, recently, researchers tried to develop automatic image analysis approaches [118][119][120][121] for time-efficient, accurate and harmonised data analysis.…”

“…Free image analysis software (e.g., MPhunter/siMPle) [122,123] and computer vision-based systems (e.g., SMACC) [118,120] can be used, respectively, on the one hand for the systematic chemical identification of MPs, and on the other hand, for the automatic counting and classification of MPs in the environment, reducing the data calculation time and the human bias during manual data analysis.…”

Deep learning approach for automatic microplastics counting and classification

“…[[Image]] The use of images taken with digital photo cameras or mobile phones has not implied that the processing time were longer than with other work using specific equipment. In a previous work (Lorenzo-Navarro et al, 2020) where SMACC system is presented, a comparison of manual versus automatic classification time is shown. In Table 3, a comparison with the proposed approach is provided.…”

“…In recent published works on microplastics classification based on image analysis, as those proposed by Lorenzo-Navarro et al (Lorenzo-Navarro et al, 2020) and Wegmayr et al (Wegmayr et al, 2020), Sauvola adaptive thresholding (Sauvola and Pietikäinen, 2000) has been applied. In order to compare the results of U-Net, the patches were segmented using the Sauvola method and the average accuracy, precision and recall was 96.39%, 71.16% and 48.37% respectively.…”

“…All the previous methods based on image analysis have required dedicated equipment as special staining dyes and light (Prata et al, 2019), flat scanner (Mukhanov et al, 2019) (Gauci et al, 2019) (Lorenzo-Navarro et al, 2020 or microscopy images (Wegmayr et al, 2020). The methodology proposed in this work is based on images taken with cameras or mobile phones with a resolution of at least 16 million pixels.…”