Bottle counting: Datasets & Models

Manual bottle counting on a food and beverage line is slow, error-prone, and hard to sustain at production speed, especially during high-throughput runs or inventory checks. Automating it with computer vision gives you a consistent count on every frame, flags miscounts before product ships, and frees staff from repetitive tallying. Because bottle detection reports the location of each unit, you can also catch missing bottles in a case or gaps on a shelf, not just totals.

Counting becomes continuous and objective, reducing the labor and rework tied to manual audits while improving fill and pack accuracy. The workflow integrates with existing line cameras and CCTV, so you can deploy without new hardware, and it scales from a single filling station to multiple lines or facilities. Using RF-DETR for detection gives strong accuracy on dense, overlapping bottle arrangements typical of beverage production.

1. Upload an image: Drop any photo into the widget above to try bottle counting for free, right on this page.
2. Preview instant results: The model detects glass bottle, plastic bottle, bottle cap, bottle crate, beverage can within seconds and returns an annotated image with structured predictions, so you can immediately see how well it works. Adjust the detection classes and re-run until the results fit your scenario.
3. Like what you see?: With a free Roboflow account, you can deploy this workflow with the Serverless Cloud API, customize it in your workspace, or train RF-DETR on your own images for production accuracy.
4. Get started for free: No code or training data required to begin: upload, review the results, and make it part of your stack in a single click.