Vérifiez instantanément les images pour détecter un contenu non sûr, estimez l'âge des personnes et obtenez une description automatique par IA directement dans votre navigateur.
ou déposez, collez ou choisissez depuis le cloud
What comes back from a run, before you upload anything.
The settings hold 3 checkboxes. Leave them as they are, or switch one off when you do not need it. At least one has to stay on, and the panel says so.
Rates the picture on a scale with 5 steps and gives a short reason for the rating. On by default.
Counts the people in the picture and returns an age range for each one. On by default.
Adds a written description of what the picture shows. Off until you switch it on.
The AI image analyzer rates every picture on a scale with 5 steps: safe, mild, moderate, risky, and highly risky. A short reason comes with the rating, so you can see why an image landed where it did.
Under the rating sits a score for each category the check looks at, among them nudity, explicit content, suggestive content, and graphic content. A borderline picture is much easier to judge when you can see which score pushed it over.
Detect explicit content is switched on when the page loads. Turn it off and the safety block comes back marked as not analyzed, never as safe, so a skipped check is never mistaken for a clean result.
Estimate age finds the people in an image and returns an age range for each one, such as 20 to 29. Every person also gets a broader age group, from baby and child through teenager and young adult to senior.
The report names the gender the analyzer detects for each person, and it raises a separate flag when someone in the picture may be a minor. The number of people found is part of the report as well.
A photo analyzer cannot see a birth certificate. Treat the numbers as a first pass that tells you which pictures a person should look at, not as a decision.
AI results are estimates and are not suitable for legal age verification or identity checks.
Describe the image is the third option. Switch it on before you start and the analyzer writes a short description of what the picture shows and adds it to the report.
A written description saves time when you need alt text for a web page, a caption for a photo library, or a plain summary of a picture you cannot open right now.
The other 2 checks are detections rather than written text. Only this option produces AI-written words, so leave it off if you want findings without prose.
Upload a picture and the same checks run every time. There is no prompt to write and no chat to steer, so 2 images come back in the same shape and can be compared line by line.
Several images can go into one run. Each image is analyzed on its own and comes back as its own result, so a batch stays as easy to read as a single file.
A free run takes up to 5 images at a time, or 10 with a free account, and each file can be up to 75 MB.
Nobody reads or copies the images you upload. Every run is automated from start to finish, so no person looks at what is inside your pictures.
The images are not used to train AI models. You keep the copyright and the ownership of the picture and of the analysis that comes back.
Processing runs on servers located in Europe, under a data processing agreement, and the download link is generated at random so that nobody can guess it.
The same run answers 3 questions at once: is this picture safe, who is in it, and what does it show.
A forum, a community site, or a marketplace can check every uploaded picture for unsafe content before it reaches a public page.
Avatars and profile photos carry the same risk as any other upload. One run gives a safety rating and a count of the people in the frame.
Switch on Describe the image and every picture comes back with a sentence you can shape into alt text or a caption.
The report says how many people are in a picture and what it shows, which is enough to group a folder of holiday photos or stock images.
Before a set of pictures goes into a slide deck, a newsletter, or a training run, one batch tells you what is actually in it.
A quick run before you post says whether a picture reads as safe, and whether it shows someone who may be a minor.
Upload any of these and the analysis runs the same way. Every cell in the last column is a typical source, not a limit.
| Format | Extension | Where it usually comes from |
|---|---|---|
| JPEG | .jpg, .jpeg | Phone cameras |
| PNG | .png | Screenshots and graphics |
| WebP | .webp | Images saved from the web |
| GIF | .gif | Short animations |
| BMP | .bmp | Older Windows software |
| TIFF | .tiff | Scans and print work |
The picture itself is never changed. The analyzer reads the image and writes a separate report, so the file you uploaded stays exactly as it was.
Analyser des images avec l'IA ne prend que quelques secondes et fonctionne directement dans votre navigateur.
Téléchargez une ou plusieurs images (JPG, PNG, WebP et plus).
Pick the checks you want in the settings panel: Detect explicit content, Estimate age, and Describe the image. At least one has to stay on.
Press Analyze and the report fills in, one section per image, with your checks applied.
Read the report for every image and download the full analysis.
An AI image analyzer is an app that looks at the picture itself instead of the data around it. This one rates an image for unsafe content, finds the people in it and estimates their age, and can write a short description of what it shows. It runs in the browser, with nothing to install.
Yes. You can analyze 5 images per run without an account, and 10 with a free account. Each image can be up to 75 MB, and no installation is needed.
Upload the photo above, leave at least one of the 3 checks switched on, and start the run. The analyzer reads the picture and hands back a report with a safety rating, the people it found, and an optional written description. Several photos can go in at once.
Yes. Estimate age returns an age range for every person it finds, such as 20 to 29, along with a broader age group and the gender it detects. It also raises a flag when someone may be a minor. These results are estimates and are not suitable for legal age verification or identity checks.
Yes. Detect explicit content rates each image on a scale with 5 steps, from safe through mild, moderate, and risky to highly risky, and gives a short reason for the rating. Scores for the individual categories sit underneath, so a borderline picture is easier to judge.
No. The analyzer describes what is in a picture and how safe it is. It does not judge whether the picture was generated by an AI model, so an image from a camera and an image from a generator are treated the same way.
JPG, JPEG, PNG, WebP, GIF, BMP, and TIFF images are all accepted. The picture itself is never changed, because the analyzer only reads it and writes a separate report.
Yes. Nobody reads or copies your images, because every run is automated. The images are not used to train AI models, processing runs on servers in Europe, and the download link is generated at random so that nobody can guess it.
Guide sur le rôle et l'importance des métadonnées dans le monde numérique
Lire l'article
Guide pour les photographes
Lire l'article
Utilisez des modèles avancés de comparaison d'images pour détecter les changements visuels et les différences de métadonnées entre deux images, idéal pour la criminalistique, la vérification d'authenticité et la recherche.
Lire l'articleEvery one of these runs in the browser, with no account and nothing to install.
