We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
LE-RTDETR: multi-scale edge enhancement and structural compression for underwater debris detection
AI summary Read the abstract
Researchers built a faster, lighter AI tool that spots trash in murky underwater photos more accurately than previous models, even with blurry visibility and distorted colors. This matters because better detection of ocean debris, including plastic waste that breaks down into microplastics, could help cleanup efforts target pollution that eventually affects marine life and the seafood we eat.
Marine debris pollution has become a major environmental concern worldwide. Underwater images often suffer from low contrast, color distortion, and weakened object boundaries due to light absorption, scattering, and complex backgrounds, making underwater debris detection particularly challenging. To address the degradation of structural information in underwater imagery and the relatively high computational cost of RT-DETR, this study proposes LE-RTDETR (Lightweight Edge-enhanced RT-DETR), a resource-efficient edge-enhanced detector for underwater objects. LE-RTDETR strengthens structural representations through controlled multi-scale edge generation and dual-stage edge–semantic fusion. Model complexity is further reduced through a multidimensional structural configuration defined prior to training, while EIoU is introduced to improve bounding-box regression. The proposed method is evaluated on three underwater debris datasets: Trash_ICRA19, Underwater Plastic Pollution Detection, and DeepTrash. On Trash_ICRA19, LE-RTDETR achieves a precision of 60.1%, a recall of 56.9%, an mAP@0.5 of 50.9%, and an mAP@0.5:0.95 of 33.0%, outperforming the baseline RT-DETR by 7.4, 9.1, 7.1, and 5.8 percentage points, respectively. Meanwhile, the number of parameters and GFLOPs are reduced by 62.3% and 64.5%, respectively, and the inference speed reaches 160.4 FPS. On the Underwater Plastic Pollution Detection and DeepTrash datasets, LE-RTDETR achieves mAP@0.5 scores of 78.5% and 85.2%, together with mAP@0.5:0.95 scores of 50.6% and 53.7%, respectively. These results demonstrate that LE-RTDETR shows consistent detection performance across different underwater debris datasets while reducing model complexity and maintaining a favorable balance between detection accuracy and inference efficiency.
More Papers Like This
Marine Debris Recognition with Artificial Intelligence
AI summary Read the abstract
Millions of tons of plastic enter our oceans every year, breaking down into tiny microplastics that work their way up the food chain and onto our plates. Researchers built an AI tool that scans underwater photos and automatically spots trash with about 82% accuracy, which could help cleanup groups and governments find and target pollution hotspots faster than manual monitoring allows. This kind of tech won't solve ocean plastic pollution on its own, but it's a practical step toward tracking the problem that ultimately threatens seafood safety and human health.
Improved Transformer-based detection of underwater plastic debris in complex environments
AI summary Read the abstract
Researchers built an improved AI system that can spot plastic trash underwater more accurately, even in murky, cluttered footage where debris is small or hard to see. This matters because better detection tools could help track and clean up ocean plastic before it breaks down into microplastics — the tiny particles increasingly found in seafood, drinking water, and even human blood and organs. While this study focused on detection technology rather than health outcomes directly, improving our ability to find and remove plastic waste is an important step toward reducing microplastic pollution that can ultimately end up in our food chain and bodies.
GoogLeNetMP: A Development of GoogLeNet Architecture for Multi-Class Microplastic Classification in Subsurface Water Image
AI summary Read the abstract
Scientists developed a smart AI tool that can automatically spot and sort microplastic pollution in underwater ocean photos with about 96% accuracy, even in murky, hard-to-see conditions. This matters because tracking microplastics in our oceans—tiny plastic bits that can end up in seafood and drinking water—is a key step toward understanding and reducing their potential risks to human health, and tools like this could make monitoring faster and cheaper than manual inspection.
Automated Underwater Plastic Waste Detection and Volume Assessment through Instance Segmentation
AI summary Read the abstract
Researchers built an AI tool that can automatically spot and measure plastic waste in murky underwater footage, even when visibility is poor. This matters because tracking ocean plastic pollution is a critical first step to cleaning it up, plastic debris breaks down into microplastics that enter the food chain and, ultimately, our own bodies through seafood and water. By making pollution monitoring faster and easier, tools like this could help scientists and cleanup crews target problem areas more efficiently.
Deep learning-based detection of submerged debris and plastics in underwater environments: a systematic review
AI summary Read the abstract
This review pulls together 50 studies on AI systems that use underwater cameras to spot plastic debris in oceans and waterways, finding that certain smart, efficient algorithms can now detect trash with over 90% accuracy—even running on small, low-power devices carried by underwater robots. This matters because better tracking of where plastic waste accumulates and breaks down is a key step toward understanding and eventually reducing the microplastics that end up in seafood, drinking water, and our bodies, though the technology still struggles in murky or dark water.
Research digests by email
When a large batch of papers lands in the Atlas, we read through it and send a short write-up of what stood out.