Turn coastal images into actionable marine litter information with AI. Upload one or more images and receive a structured detection report. Use JPG/PNG images where litter objects are clearly visible.
Set a global confidence threshold for all images or set it for each image individually using the threshold field on its card.
Example Analysis Result
Report
Works
Upload one or more images of coastal areas.
AI model detects and localizes visible litter items.
Items are classified based on our marine litter taxonomy.
Get a summary and download your detailed report.
Summary
In case of failed images, you can click on them to retry running them
Total Images: -
Completed Images: -
Failed Images: -
Currently Processing: -
Total Litter Objects Detected: -
Litter Objects Detected per Image: -
Average Confidence Threshold: -
Average Detection Confidence: -
No detected materials yet.
No detected use/function categories yet.
| Specific Type | Material | Use / Function | Count |
|---|
No specific litter types detected yet.
Total Objects Detected: -
Confidence Threshold: -
Average Confidence: -
No detected materials yet.
No detected use/function categories yet.
| Specific Type | Material | Use / Function | Count |
|---|
No specific litter types detected yet.
Archive
WorksLaskAI turns coastal images into reliable marine litter information through a simple 4-step process powered by advanced AI.
Upload one or more images of coastal environments.
Supported formats: JPG, PNG
Multiple images at onceThe AI model detects and localizes visible litter objects in the images.
State-of-the-art object detection model
High accuracy localization
Works in diverse conditionsDetected items are classified using our marine litter taxonomy.
Plastic Drink BottleConfidence: 76%
Plastic Drink BottleConfidence: 73%
Glass BottleConfidence: 69%
Plastic Non-Food ContainerConfidence: 61%
Paper BagConfidence: 57%
Hierarchical classification:
59 specific litter categories
Consistent and comparable resultsGet a summary of the findings and download your detailed report.
By Material
By Use / Function
LaskAI uses a structured, hierarchical taxonomy to ensure accuracy and consistency.
Level 1: Material
Level 2: Use / Function
Level 3: Specific Item Type
… and more across all 7 materials, 13 use/function groups, and 59 item types.
Terms of Use and Privacy Notice
Last updated: 15 September 2026
LaskAI is a web application for uploading coastal images, detecting and classifying visible marine litter, and creating reports. By using LaskAI, you agree to use it responsibly and in accordance with these terms.
Analysis results are generated automatically and are provided for information and monitoring purposes. They may contain errors or omissions and should be checked before being used for decisions, reports, or research.
You may upload only images that you own or are authorised to use. You keep all rights in your images. You give Scidrones permission to store, process, and display them only as needed to provide the analysis, archive, and reports you request.
Do not upload unlawful content, malicious files, or material that infringes another person's rights. Avoid uploading identifiable people, vehicle registration plates, private documents, sensitive information, or unnecessary location metadata. Do not attempt to access another user's information, bypass security measures, or disrupt the service.
Keep your password private and use an email address you control. You are responsible for activity carried out through your account. Scidrones may limit or suspend access where it is reasonably necessary to protect the service, its users, or applicable law.
The service may be updated, changed, or temporarily unavailable for maintenance or technical reasons. These terms may be updated from time to time; the latest version will be available in the application.
Controller and contact: Scidrones operates LaskAI. You can contact Scidrones at info@scidrones.com or at Evangelistrias 30, Mytilene 81100, Greece.
Information we process: this includes account details such as username and email address; password hashes and session, verification, and security information; uploaded images, filenames and image metadata; analysis settings, results, annotated images, and generated reports. Images and their metadata may contain personal data, including location information.
Why we use it: we use this information to provide the service you request, operate accounts and archives, generate reports, prevent abuse, maintain security, and respond to support or privacy requests. Scidrones does not use uploaded images for advertising, sale, publication, research, or model training.
Authorised administrators and service providers that support hosting, email, security, and delivery of the service may process information only as needed to provide their services. Cloudflare Turnstile processes technical signals for bot protection; see the Cloudflare Turnstile Privacy Notice.
LaskAI uses a session cookie for login and guest-session continuity. It also uses browser storage for temporary session and upload references and to remember your theme preference. Clearing this browser storage may interrupt these features but does not delete server-side uploads.
Uploads and results remain available in your archive until you delete them. You can delete uploads from the archive and request account deletion through your profile. Deleting an account permanently removes all associated account records; stored files will be removed through server cleanup rather than immediately.
Where the GDPR applies, you may request access to, correction, erasure, restriction, or portability of your personal data, and may object to processing where the relevant conditions apply. To make a request, email info@scidrones.com. You may also complain to your local data-protection authority. In Greece, this is the Hellenic Data Protection Authority.
Enter the six-digit code sent to your email.
Enter the email address connected to your account.
Enter the six-digit code sent to your email.
LaskAI: Marine Litter Detector and Reporter is an AI-powered web application designed to support the detection, classification, and reporting of marine litter from images.
Users can upload images of coastal environments, and LaskAI automatically analyses them to identify visible litter items, classify them according to a structured marine litter taxonomy, and generate an easy-to-understand report of the detected items.
At the core of LaskAI is a dedicated machine-learning model developed specifically for marine litter recognition. The classification framework follows a hierarchical approach based on material, use/function, and specific litter type, enabling consistent and detailed reporting across different coastal environments. LaskAI_Classification.xlsx
LaskAI was funded by the A.C. Laskaridis Charitable Foundation and developed by Scidrones, with the aim of making advanced artificial intelligence tools for marine litter monitoring more accessible and easier to use.
By transforming ordinary images into structured information, LaskAI provides a simple bridge between citizen observations, artificial intelligence, and marine litter monitoring, supporting a better understanding of the presence and composition of litter in our coastal environments.
By using the LaskAI web application, you accept and comply with the terms of use and policy.
Admin
Open this page to load active sessions.
Open this page to load user accounts.
Confirmation
Marine litter taxonomy
