Release Notes | NuditySearch Release to Mashape| NSP 1.2.15.1

This is a planned update to NuditySearch. The release contains updates to nudity detection to both reduce false positives and increase accuracy. The release number has been incremented to reflect changes to the core software.

Products Affected

Update to NuditySearch NSP 1.2.6.1

New Features

Improvements to Classification Engine

A multi-layer classification process has been implemented to provide for both a specific classification of genitals, breasts, and buttocks, as well as a general classification of nudity when sexually suggestive body regions are displayed.

Environmental Awareness

On the Loose and Standard sensitivity setting, the engine now takes into account the environment and surroundings of the subject matter. Because of the fine line between lingerie and bathing suits, intent can become an important perspective. By taking into account the surrounding environment, our classification process now includes more situational awareness to reach a conclusion.

Sensitivity parameter now called Setting

ImageVision allows customers to define their tolerance for nudity vs false positives by adjusting the solution sensitivity settings to fit your application of the technology.

There is an explicit trade off related to the definition of nudity and the ability for any computer technology to recognize it. When a strict definition is applied, the technology is more likely to identify borderline cases as nudity or non-nude images as nudity making a false positive identification. With a more relaxed setting, images may be misidentified as “clean” when in fact; there is nude imagery or a false negative identification.

The different sensitivity settings allow for different degrees of false negative and false positive tolerances of the solution allowing for the definition of nudity to fit the application of the technology.

After studying usage patterns over the past few months, we are simplifying the engine sensitivity settings to three options – Loose, Standard, and Strict. To maintain backwards compatibility, we are going to keep the Sensitivity parameter values 1 – 10, but map them to a new parameter called Setting. Sensitivity 1 – 3 are mapped to Setting 1 (Loose). Sensitivity 4-7 are mapped to Setting 2 (Standard). Sensitivity 8-10 are mapped to Setting 3 (Strict).


Release Notes | NuditySearch Release to Mashape| NSP 1.2.6.1

This is a planned update to NuditySearch. The release contains updates to nudity detection to improve performance. The release number has been incremented to reflect changes to the core software.

Products Affected

Update to NuditySearch NSP 1.2.5.1

New Features

Improvements to Texture Analysis

Texture is a key component of human visual perception. Like color, this makes it an essential feature to consider when querying image databases. Everyone can recognize texture but it is more difficult to define. Unlike color, texture occurs over a region rather than at a point. It is normally defined purely by grey levels and as such is orthogonal to color. Texture has qualities such as periodicity and scale; it can be described in terms of direction, coarseness, contrast and so on. ImageVision improved the texture analysis in the following attributes:

Coarseness
Contrast Directionality
Line-likeness
Regularity
Roughness

These attributes allow us to better categorize skin, and non-skin regions such as natural surfaces (i.e. sand, earth, sky, wood and water). Allowing for better contextual decision making regarding the image.

Sensitivity Settings

ImageVision allows customers to define nudity by adjusting the solution sensitivity settings so that a specific definition can drive the application of the technology.

There is an explicit trade off related to the definition of nudity and the ability for any computer technology to recognize it. When a strict definition is applied, the technology is more likely to identify borderline cases as nudity or non-nude images as nudity making a false positive identification. With a more relaxed setting, images may be misidentified as “clean” when in fact; there is nude imagery or a false negative identification.

The different sensitivity settings allow for different degrees of false negative and false positive tolerances of the solution allowing for the definition of nudity to fit the application of the technology.

ImageVision recommends setting for those seeking to achieve the least false negatives.


Release Notes | NuditySearch Release to Mashape

Release Notes for ImageVision NuditySearch Release NSP1.2.5.1 - Major Release Mashape Configuration 28.7101

Products Affected:

Mashape version NuditySearch has been upgraded to 1.2.5.1, config. 28.7101

Overview:

ImageVision has developed an automated visual recognition platform with unique detection modules that detect and recognize objects based upon computer vision technology. Visual information is accurately classified allowing the solution to search for visual content or to filter out visual content based upon the configuration of the platform. Once the content is categorized, it can be managed by policies set forth by developers.

Nudity recognition is a complex problem due to variability in both the definition and subject matter. Regardless, few want pornography, explicit nude images and the like, on their webpages or social networks.

There is an explicit trade off related to the definition of nudity and the ability of any computer technology to recognize it. When a strict definition is applied, the technology is more likely to identify borderline cases as nudity or non-nude images as nudity making a false positive identification. With a more relaxed setting, images may be misidentified as “clean” when in fact; there are adult themes represented or a false negative identification.

ImageVision’s NuditySearch allows customers to define nudity by adjusting the solution sensitivity settings so that a specific definition can drive the application of the technology.

Different sensitivity settings allow for different degrees of exposure tolerances of sexually identifiable body parts. At the default setting, ImageVision's technology attempts to accurately identify skin and exposed sexually identifiable body parts. To do this, ImageVision's proprietary research focuses on skin texture recognition, edge extraction to discern patterns and shapes, and artificial intelligence reasoning based upon complex, multi-layered decision tree.

Feature Improvements:

  1. General Skin Map Change:

1.1. Improved segmentation analysis to filter out non-skin textured surfaces. 1.2. Identify threshold for surface analysis so that a segment of the photo area can be eliminated, or determined to have no skin. Eliminates textured foreground and background. 1.3. Improves on false skin detection issues using improved texture and shape detection. Improve “negative” non-skin textures from images. 1.4. Identified Parallel line textures in hardwood and tile to eliminate false skin detection

  1. Close-up Hands and Feet Detection:

2.1. Enclosed segment analysis: 2.2. Identifies encapsulated segments in low quality images (Faces, Eyes, Nose, Hair). Improves False Positives rates.

  1. Face & Head Detection:

3.1. A more robust face detection algorithm for detecting obscured and partially exposed faces. 3.2. Improves performance on very close up face images. 3.3. Improved Close-up Face detection with images with profile faces and multiple faces.

  1. Improve False Positives on Images of Babies & Extreme Close-Ups:

4.1. Better detection of Baby Bodies 4.2. Removes False Positives of Close-Up Lips, Mouth-Only

Last 7 days, UTC

Average Latency
38ms

Average Uptime
100.0%

Current Status
Online

  • No information
  • Outage
  • Disruption
  • Normal
Average latency determined from Mashape to API Response time

Simple & Straightforward Pricing

Pay as you go. No long-term contracts.

Freemium

$0

additional fees may apply

Image Volume

250 / day

$0.0025 per extra

Subscribe

premium

$399.00

per month

Images

10,000 / day

$0.0013 per extra

Subscribe

ultra

$599.00

per month

Images

20,000 / day

$0.0010 per extra

Subscribe

Endpoints

HTTP 200 : Nudity Search Response POST/nuditySearch/image
						{
  "version": "config_v8_2014-01-31",
  "message": "SUCCESS",
  "transactionid": "1234",
  "objecturl": "http://i.imgur.com/4hGni1I.jpg",
  "setting": 2,
  "score": 94,
  "classification": "NUDITY"
}						
NuditySearch

Use this API to detect nudity in images. Nobody gets more naked! NOTE: API accepts JSON and XML encoded posts.


Test console
Description
Parameter
string

Image to scan for nudity
Example: http://i.imgur.com/4hGni1I.jpg

objecturl

Required

number

Deprecated - Please use 'Setting' parameter. Sensitivity 1 – 3 are mapped to Setting 1 (Loose). Sensitivity 4-7 are mapped to Setting 2 (Standard). Sensitivity 8-10 are mapped to Setting 3 (Strict).
Example: 6

sensitivity

Optional

number

Optional scan sensitivity setting - Valid settings are 1 (Loose), 2 (Standard) and 3 (Strict)
Example: 2

setting

Optional

You must have an API key to test this API!


Mashape allows developers to find, consume, and distribute cloud APIs just like ImageVision - NuditySearch.

Login to your account or signup: Create Account

or
   Signup with GitHub

By signing up you agree to our terms of service.