Light

pauloprado / pdi Goto Github PK

View Code? Open in Web Editor NEW

0.0 1.0 0.0 14.02 MB

Python 100.00%

pdi's Introduction

pdi

Usage: python indexExtractor -i list.txt -f [filterType] -o outputdir

Where the list.txt must have the img and the ground truth in the same line.

[filterType]:

0 -> No filter

1 -> Blur

2 -> Gaussian filter

3 -> Mean filter

4 -> Bilateral filter

No filter results

Method	AUC	EER	FAR	FRR	Accuracy
NGRDI	0.834	0.285	0.261	0.260	0.739
ExG	0.802	0.224	0.288	0.288	0.712
CIVE	0.856	0.576	0.224	0.224	0.776
VEG	0.798	0.193	0.285	0.285	0.715
ExGR	0.826	0.226	0.266	0.266	0.734
WI	0.797	0.425	0.264	0.264	0.736

Method	AUC	EER	FAR	FRR	Accuracy
Arithmetic Mean	0.857	0.325	0.230	0.230	0.770
Geometric mean	0.852	0.008	0.235	0.235	0.765

Method	Accuracy
Majority	0.740

Blur results

Method	AUC	EER	FAR	FRR	Accuracy
NGRDI	0.899	0.386	0.190	0.190	0.810
ExG	0.858	0.314	0.236	0.236	0.764
CIVE	0.882	0.604	0.198	0.198	0.802
VEG	0.855	0.288	0.240	0.240	0.760
ExGR	0.889	0.326	0.201	0.201	0.799
WI	0.817	0.428	0.256	0.256	0.744

Method	AUC	EER	FAR	FRR	Accuracy
Arithmetic Mean	0.894	0.395	0.197	0.197	0.803
Geometric mean	0.893	0.067	0.198	0.198	0.802

Method	Accuracy
Majority	0.778

Gaussian filter results

Method	AUC	EER	FAR	FRR	Accuracy
NGRDI	0.879	0.349	0.210	0.210	0.790
ExG	0.837	0.284	0.254	0.254	0.746
CIVE	0.877	0.597	0.205	0.205	0.795
VEG	0.836	0.257	0.256	0.256	0.744
ExGR	0.865	0.288	0.226	0.226	0.774
WI	0.803	0.421	0.264	0.264	0.736

Method	AUC	EER	FAR	FRR	Accuracy
Arithmetic Mean	0.879	0.371	0.210	0.210	0.790
Geometric mean	0.877	0.035	0.211	0.211	0.789

Method	Accuracy
Majority	0.763

Mean filter results

Method	AUC	EER	FAR	FRR	Accuracy
NGRDI	0.889	0.387	0.202	0.202	0.798
ExG	0.834	0.302	0.263	0.263	0.737
CIVE	0.891	0.604	0.193	0.193	0.807
VEG	0.814	0.268	0.282	0.282	0.718
ExGR	0.865	0.319	0.231	0.231	0.769
WI	0.805	0.429	0.267	0.267	0.733

Method	AUC	EER	FAR	FRR	Accuracy
Arithmetic Mean	0.881	0.390	0.214	0.214	0.786
Geometric mean	0.879	0.057	0.215	0.215	0.785

Method	Accuracy
Majority	0.758

Bilateral filter results

Method	AUC	EER	FAR	FRR	Accuracy
NGRDI	0.839	0.289	0.256	0.256	0.744
ExG	0.806	0.231	0.285	0.285	0.715
CIVE	0.870	0.588	0.213	0.213	0.787
VEG	0.795	0.203	0.291	0.291	0.709
ExGR	0.827	0.230	0.267	0.267	0.733
WI	0.796	0.419	0.266	0.266	0.734

Method	AUC	EER	FAR	FRR	Accuracy
Arithmetic Mean	0.862	0.331	0.226	0.226	0.774
Geometric mean	0.856	0.010	0.232	0.232	0.768

Method	Accuracy
Majority	0.739

Original	Early Fusion

Blur	Early Fusion

Gaussian	Early Fusion

Mean	Early Fusion

Bilateral	Early Fusion

Database LCRS2

No filter results

Method	AUC	EER	FAR	FRR	Accuracy
NGRDI	0.790	0.428	0.290	0.290	0.710
ExG	0.801	0.311	0.268	0.268	0.732
CIVE	0.838	0.559	0.236	0.236	0.764
VEG	0.783	0.274	0.289	0.289	0.711
ExGR	0.777	0.374	0.302	0.302	0.698
WI	0.835	0.399	0.231	0.231	0.769

Method	AUC	EER	FAR	FRR	Accuracy
Arithmetic Mean	0.842	0.391	0.241	0.241	0.759
Geometric mean	0.839	0.058	0.244	0.244	0.756

Method	Accuracy
Majority	0.751

Blur results

Method	AUC	EER	FAR	FRR	Accuracy
NGRDI	0.865	0.509	0.209	0.209	0.791
ExG	0.861	0.388	0.215	0.215	0.785
CIVE	0.853	0.573	0.220	0.220	0.780
VEG	0.853	0.367	0.225	0.225	0.775
ExGR	0.860	0.462	0.212	0.212	0.788
WI	0.813	0.418	0.252	0.252	0.748

Method	AUC	EER	FAR	FRR	Accuracy
Arithmetic Mean	0.881	0.457	0.193	0.193	0.807
Geometric mean	0.882	0.261	0.193	0.193	0.807

Method	Accuracy
Majority	0.801

Gaussian filter results

Method	AUC	EER	FAR	FRR	Accuracy
NGRDI	0.849	0.480	0.232	0.232	0.768
ExG	0.849	0.361	0.230	0.230	0.770
CIVE	0.848	0.573	0.225	0.225	0.775
VEG	0.840	0.338	0.238	0.238	0.762
ExGR	0.840	0.432	0.240	0.240	0.760
WI	0.820	0.409	0.247	0.247	0.753

Method	AUC	EER	FAR	FRR	Accuracy
Arithmetic Mean	0.874	0.436	0.204	0.204	0.796
Geometric mean	0.874	0.169	0.204	0.204	0.796

Method	Accuracy
Majority	0.789

Mean filter results

Method	AUC	EER	FAR	FRR	Accuracy
NGRDI	0.859	0.502	0.215	0.215	0.785
ExG	0.852	0.391	0.224	0.224	0.776
CIVE	0.851	0.584	0.223	0.223	0.777
VEG	0.838	0.362	0.238	0.238	0.762
ExGR	0.845	0.454	0.226	0.226	0.774
WI	0.826	0.426	0.240	0.240	0.760

Method	AUC	EER	FAR	FRR	Accuracy
Arithmetic Mean	0.876	0.456	0.199	0.199	0.801
Geometric mean	0.876	0.255	0.199	0.199	0.801

Method	Accuracy
Majority	0.796

Bilateral filter results

Method	AUC	EER	FAR	FRR	Accuracy
NGRDI	0.806	0.437	0.280	0.280	0.720
ExG	0.813	0.322	0.262	0.262	0.738
CIVE	0.843	0.567	0.230	0.230	0.770
VEG	0.796	0.288	0.282	0.282	0.718
ExGR	0.793	0.380	0.293	0.293	0.707
WI	0.830	0.399	0.239	0.239	0.761

Method	AUC	EER	FAR	FRR	Accuracy
Arithmetic Mean	0.850	0.401	0.235	0.235	0.765
Geometric mean	0.847	0.074	0.237	0.237	0.763

Method	Accuracy
Majority	0.756

Original	Early Fusion

Blur	Early Fusion

Gaussian	Early Fusion

Mean	Early Fusion

Bilateral	Early Fusion

Indices comparison blurred input

Original

NGRDI	CIVE	ExG

ExGR	VEG	WI

Database 1 - Individual sample analysis

	Mean	Standard deviation
AUC	0.9528	0.02942
Accuracy	0.891429	0.042455

Database 2 - Individual sample analysis

	Mean	Standard deviation
AUC	0.912412	0.055023
Accuracy	0.848118	0.061750

BoxPlot

Data	NGRDI	NGRDI + Blur	CIVE	CIVE + Blur
AUC
EER
AUC
EER

Best and worst result

NGRDI	NGRDI + Blur	CIVE	CIVE + Blur

pdi's People

Contributors

Watchers

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.