← Latest papers
🌿 ecology

Rapid estimation of insect biodiversity using AI assisted image analysis:a cost-effective tool for monitoring and ecological assessmen

This study demonstrates that an AI-assisted image analysis tool can rapidly and cost-effectively estimate flying insect biodiversity by using visual morphological features as a reliable proxy for taxonomic diversity, thereby overcoming the need for extensive expert identification in ecological monitoring.

Original authors: Cabra-Leykin, L., Keasar, T., Keasar, C.

Published 2026-11-13
📖 5 min read🧠 Deep dive

Original authors: Cabra-Leykin, L., Keasar, T., Keasar, C.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

The health of our planet depends heavily on the tiny, often overlooked creatures that inhabit it. Insects, for instance, are the unsung engineers of ecosystems; they pollinate our food crops, recycle nutrients, and serve as a food source for countless other animals. When insect populations decline or change, it is a warning signal that the environment is under stress. However, keeping a close watch on these populations is incredibly difficult. To understand what is happening, scientists traditionally have to catch insects, identify them down to their specific species, and count them. This process is slow, expensive, and requires a level of specialized knowledge that is becoming increasingly rare. As a result, we often lack the timely, large-scale data needed to make smart decisions about protecting nature and managing our landscapes.

A team of researchers in Israel has developed a new way to bypass this bottleneck, using a combination of simple traps and artificial intelligence to estimate insect diversity without needing to name every single bug. Instead of asking a human expert to sort through thousands of tiny specimens, the team captured flying insects on yellow sticky traps, scanned the traps like documents, and fed the images into a computer program. The software does not try to identify the species; instead, it measures the visual variety of the bugs on the trap, looking at differences in their size, shape, and color. The researchers found that this visual variety acts as a reliable mirror for the actual biological diversity of the insects. When the computer saw a wide mix of different-looking bugs, the actual number of different insect types was also high. When the trap was crowded with many bugs that all looked very similar, the biological diversity was low.

The study, conducted across thirty sites in Israel's Mediterranean region, involved setting up yellow sticky traps in both natural landscapes and agricultural fields. These traps, which are essentially sheets of adhesive coated in a bright yellow color, were placed on trees or poles and left for ten days to catch flying insects. Once collected, the researchers covered the traps with a clear sheet to protect the specimens and scanned them at a very high resolution using standard office scanners. This created a massive digital library of insect images. The team then used a computer model to locate every insect on the scans and cut them out as individual images. From these images, the computer extracted five key visual features: the size of the insect, the ratio of its length to its width, and the balance of red, green, and blue colors in its body.

By plotting these visual features, the researchers created a digital map where each insect is a point. In this map, insects that look similar cluster together, while those that look different spread out. The team developed a method to count how many distinct "clusters" or groups of visual types existed on each trap. They called this the visual diversity index. To check if this method worked, they compared the computer's estimates against the results of a traditional expert analysis. A group of professional entomologists manually identified over ten thousand insects from a selection of the traps, sorting them into families and orders. The results showed a strong agreement between the two methods. The computer's visual diversity score matched the expert's taxonomic diversity score very closely, proving that the variety of shapes and colors on a trap is a good stand-in for the variety of species.

The researchers applied this new tool to the entire dataset of 322 traps to answer some big questions about how insects live in different environments. They discovered a surprising relationship between the number of insects and their diversity. In traps where the insects were extremely abundant, the diversity was actually lower. This suggests that when an environment is teeming with bugs, it is often because a few dominant species have taken over, crowding out the others. In contrast, traps with fewer insects often held a more balanced mix of different types. The study also revealed how insect communities change between seasons and between natural and agricultural lands. In natural areas, the diversity of insects increased significantly from spring to summer, while in agricultural fields, the diversity remained more stable. This stability in the farms was likely due to irrigation and the constant presence of trees, which provided a steady environment for insects even as the natural landscape dried out in the heat of summer.

This approach offers a practical solution to the "taxonomic impediment," the global shortage of experts needed to identify insects. By using a low-cost method that relies on sticky traps and office scanners, the researchers can survey vast areas quickly and cheaply. The computer analysis is fast, saving hours of manual work for every single trap. While the method does not tell scientists exactly which species are present, it provides a rapid, standardized way to monitor the health of insect populations. This is crucial for conservation efforts, as it allows land managers and policymakers to see changes in biodiversity in real-time. The researchers emphasize that this tool is not meant to replace human experts but to expand their capacity, handling the routine screening of bulk samples so that experts can focus on deeper analysis. By turning images into data, this study provides a scalable path toward understanding and protecting the invisible world of insects that sustains our own.

Drowning in papers in your field?

Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.

Try Digest →