False Positive Rate (FPR)
Definition
The proportion of true non-matches (or true negatives) that a classification method declares as matches (or positives). Also called the type I error rate or 1 minus specificity. FPR = FP / (FP + TN).
Related terms
- Blind Steganalysis
- Detection without prior knowledge of the embedding algorithm. A classifier trained on images with and without payloads generalises across multiple tools.
- Decision Threshold
- The cut-off score above which a classification method declares a positive result. Raising the threshold reduces FPR but increases FNR; lowering it...
- Embedding Rate
- The payload size divided by the carrier capacity, usually expressed as bits per pixel. Detection difficulty decreases sharply at low embedding rates...
- False Negative Rate (FNR)
- The proportion of true matches (or true positives) that a classification method fails to detect, declaring them as non-matches. Also called the...
- Rich Model Steganalysis (SRM)
- A feature-extraction approach that computes hundreds of statistical features from pixel-residual co-occurrence matrices and feeds them to an ensemble classifier such as...
- ROC Curve
- Receiver operating characteristic curve: a plot of sensitivity (y-axis) against FPR (x-axis) as the decision threshold is swept from its most lenient...
- Sensitivity (True Positive Rate)
- The proportion of true matches correctly identified as matches. Sensitivity = TP / (TP + FN) = 1 minus FNR. A method...
- Specificity (True Negative Rate)
- The proportion of true non-matches correctly identified as non-matches. Specificity = TN / (TN + FP) = 1 minus FPR. A method...
- Steganalysis
- The forensic discipline of detecting the presence of hidden data in a carrier file. Steganalysis uses statistical tests (chi-squared, RS analysis, sample...
- Targeted Steganalysis
- Detection methods designed against a specific steganography tool or algorithm. Effective when the tool is known but fails against novel or unknown...
Explained in these topics
- False Positive and False Negative Error RatesThe proportion of true non-matches (or true negatives) that a classification method declares as matches (or positives). Also called the type I error rate or 1...
- Steganalysis: Statistical Detection MethodsThe proportion of clean (non-stego) files that a detector incorrectly flags as containing hidden data. High FPR increases investigator workload; it does not by...