[DEVELOPER TOOLS · AUTOMATION]

OpenCV Image Analyzer: Faces, Quality & Colors

Analyze image URLs with OpenCV: detect faces with landmarks, score blur and exposure, extract color palettes and hashes. Export CSV, JSON, Excel.

Input: 9 fields, required: imageUrls. Billed per event on the free Apify plan: Image analyzed $0.015, Face detection (YuNet) $0.01, Quality metrics $0.01 and 2 more events. Results export as JSON, CSV or Excel. Last updated 2026-10-04.

Run on Apify Need a custom version

STATUS: PRODUCTION
INPUT FIELDS: 9
REQUIRED: 1
PRICING: PAY PER EVENT
UPDATED: 2026-10-04
OUTPUT: JSON, CSV, Excel

01 / INPUT

What you can configure.

FIELDTYPEDESCRIPTION

Image URLsimageUrlsREQUIREDarrayDirect image URLs to analyze with OpenCV, one per line or imported from a file. Supports JPEG, PNG, WebP, GIF (first frame), AVIF, TIFF and SVG up to 30 MB each. Each valid image becomes one dataset row with faces, quality, colors and hashes. default: [{"url":"https://raw.githubusercontent.com/opencv/opencv/4.x/samples/data/lena.j

Max ItemsmaxItemsintegerMaximum number of images to analyze and bill per run (failed downloads do not count). Free users: limited to 10 images (preview). Paid users: up to 1,000,000. default: 10

Detect faces (YuNet DNN)detectFacesbooleanDetect faces with OpenCV's YuNet deep-learning detector: bounding box, confidence score and 5 facial landmarks (eyes, nose, mouth corners) per face. default: true

Min face confidence (%)minFaceConfidenceintegerFaces below this confidence score are discarded. 60 is a good default; lower it to catch small/partial faces, raise it to keep only frontal, clear faces. default: 60

Quality metrics (blur, exposure)qualityMetricsbooleanSharpness score (variance of Laplacian), blurry yes/no, brightness, contrast, exposure verdict and clipped shadows/highlights percentages. default: true

Dominant colors (k-means)extractColorsbooleanExtract the dominant color palette via k-means clustering: hex code and coverage percentage per color, plus the average color. default: true

Palette sizecolorCountintegerHow many dominant colors to extract per image (k in k-means). default: 5

Perceptual hashes (aHash + dHash)perceptualHashboolean64-bit average hash and difference hash for near-duplicate detection. Images with a small Hamming distance between hashes are visually similar. default: true

Proxy configurationproxyConfigurationobjectMost image CDNs are open, so no proxy is needed by default. Enable Apify Proxy only if the image host blocks datacenter IPs. default: {"useApifyProxy":false}

02 / PRICING

Pay for what you collect.

EVENTDETAILPRICE

Image analyzed · BASEBase analysis row per image: format, dimensions, megapixels, aspect ratio and file size.$0.015

Face detection (YuNet)Deep-learning face detection: bounding box, confidence and 5 landmarks per face.$0.01

Quality metricsSharpness, blur verdict, brightness, contrast, exposure and clipping percentages.$0.01

Dominant colorsK-means color palette with hex codes and coverage, plus average color.$0.01

Perceptual hashes64-bit aHash and dHash for near-duplicate detection.$0.01

Prices in USD, shown for the Apify free plan. Paid Apify plans get tiered discounts. Platform usage is included.

03 / FAQ

About the OpenCV Image Analyzer.

How much does the OpenCV Image Analyzer cost?

It charges per event: Image analyzed $0.015, Face detection (YuNet) $0.01, Quality metrics $0.01 and 2 more events. Prices are for the free Apify plan, and paid Apify plans get tiered discounts. Apify platform usage is included in the event price.

What input does the OpenCV Image Analyzer need?

One field is required: imageUrls (array). The Actor has 9 input fields in total, and the rest have defaults.

Which formats can I export the OpenCV Image Analyzer results in?

Results are saved to an Apify dataset. Download them as JSON, CSV or Excel, or fetch them through the Apify API.

04 / RELATED