Phase congruency based on derivatives of circular symmetric Gaussian function: an efficient feature map for image quality assessment

Chen, Congmin; Mou, Xuanqin

doi:10.1186/s13640-023-00611-2

EURASIP Journal on Image and Video Processing

Table 3 Performances of proposed metric and competitors on each individual distortion type in terms of SROCC. The top three algorithms are presented in boldface for each distortion type

From: Phase congruency based on derivatives of circular symmetric Gaussian function: an efficient feature map for image quality assessment

	Distortion	SSIM	MS-SSIM	IW-SSIM	IFC	VIF	FSIM	FSIMc	GMSD	\({q}_{m,g}\)	\({q}_{\mathrm{sd},g}\)	\({q}_{m,c}\)	\({q}_{\mathrm{sd},c}\)
LIVE	JP2K	0.9614	0.9654	0.9653	0.9100	0.9683	0.9717	0.9724	0.9711	0.9672	0.9707	0.9659	0.9710
	JPEG	0.9764	0.9793	0.9809	0.9440	0.9842	0.9834	0.9840	0.9782	0.9801	0.9797	0.9800	0.9795
	WN	0.9694	0.9731	0.9671	0.9377	0.9845	0.9652	0.9716	0.9737	0.9494	0.9624	0.9614	0.9584
	GB	0.9517	0.9584	0.9722	0.9649	0.9722	0.9708	0.9708	0.9567	0.9539	0.9662	0.9535	0.9662
	FF	0.9556	0.9321	0.9443	0.9644	0.9652	0.9499	0.9519	0.9416	0.9558	0.9578	0.9590	0.9606
CSIQ	AWN	0.8974	0.9471	0.9377	0.8460	0.9571	0.9262	0.9359	0.9676	0.9507	0.9637	0.9577	0.9644
	JPEG	0.9546	0.9622	0.9664	0.9395	0.9705	0.9654	0.9664	0.9651	0.9603	0.9652	0.9616	0.9659
	JP2K	09606	0.9691	0.9681	0.9262	0.9672	0.9685	0.9704	0.9717	0.9577	0.9715	0.9603	0.9717
	PGN	0.8922	0.9330	0.9057	0.8279	0.9509	0.9234	0.9370	0.9502	0.9317	0.9432	0.9476	0.9449
	GB	0.9609	0.9720	0.9781	0.9593	0.9747	0.9729	0.9729	0.9712	0.9535	0.9710	0.9537	0.9711
	Contrast	0.7922	0.9521	0.9540	0.5416	0.9361	0.9420	0.9438	0.9040	0.9368	0.9374	0.9414	0.9360
TID2013	AWGN	0.8671	0.8645	0.8438	0.6611	0.8994	0.8973	0.9101	0.9462	0.9033	0.9352	0.9154	0.9355
	ANMC	0.7726	0.7729	0.7514	0.5351	0.8299	0.8208	0.8537	0.8684	0.8276	0.8555	0.8583	0.8680
	SCN	0.8515	0.8543	0.8166	0.6601	0.8834	0.8750	0.8900	0.9350	0.8846	0.9251	0.9015	0.9250
	MN	0.7767	0.8014	0.8063	0.6732	0.8642	0.7944	0.8094	0.7075	0.7937	0.7547	0.8173	0.7650
	HFN	0.8634	0.8603	0.8553	0.7405	0.8972	0.8984	0.9040	0.9162	0.8879	0.9071	0.8994	0.9085
	IMN	0.7503	0.7628	0.7281	0.6407	0.8536	0.8072	0.8251	0.7637	0.7893	0.7236	0.8074	0.7302
	QN	0.8657	0.8705	0.8467	0.6282	0.7853	0.8719	0.8807	0.9049	0.8606	0.9013	0.8707	0.9014
	GB	0.9667	0.9672	0.9701	0.8906	0.9649	0.9550	0.9551	0.9113	0.9675	0.9521	0.9672	0.9518
	DEN	0.9254	0.9267	0.9152	0.7779	0.8910	0.9301	0.9330	0.9525	0.9317	0.9440	0.9346	0.9439
	JPEG	0.9200	0.9265	0.9186	0.8356	0.9191	0.9324	0.9339	0.9507	0.9351	0.9467	0.9397	0.9475
	JP2K	0.9468	0.9504	0.9506	0.9077	0.9516	0.9577	0.9589	0.9657	0.9569	0.9626	0.9598	0.9624
	JGTE	0.8493	0.8475	0.8387	0.7425	0.8409	0.8464	0.8610	0.8403	0.8647	0.8600	0.8820	0.8637
	J2TE	0.8828	0.8888	0.8656	0.7769	0.8760	0.8913	0.8919	0.9136	0.9007	0.9078	0.9016	0.9075
	NEPN	0.7821	0.7968	0.8010	0.5736	0.7719	0.7917	0.7937	0.8140	0.8055	0.8258	0.8083	0.8271
	Block	0.5720	0.4800	0.3716	0.2413	0.5306	0.5489	0.5532	0.6625	0.6480	0.6638	0.6506	0.6698
	Mean shift	0.7752	0.7906	0.7833	0.5522	0.6275	0.7531	0.7487	0.7351	0.7180	0.7162	0.7101	0.7170
	Contrast	0.3775	0.4633	0.4592	− 0.180	0.8385	0.4686	0.4679	0.3235	0.4843	0.3256	0.4838	0.3229
	CCS	− 0.414	− 0.410	− 0.420	− 0.403	− 0.310	− 0.275	0.8359	− 0.295	− 0.396	− 0.316	0.8293	0.8379
	MGN	0.7803	0.7785	0.7727	0.6142	0.8468	0.8469	0.8569	0.8886	0.8189	0.8612	0.8401	0.8626
	CN	0.8566	0.8527	0.8761	0.8160	0.8946	0.9121	0.9135	0.9298	0.8928	0.9206	0.8963	0.9214
	LCN	0.9057	0.9067	0.9037	0.8180	0.9203	0.9466	0.9485	0.9629	0.9416	0.9630	0.9435	0.9632
	CQD	0.8542	0.8554	0.8401	0.6006	0.8414	0.8760	0.8815	0.9102	0.8867	0.9059	0.8940	0.9069
	Chr. abr	0.8775	0.8784	0.8681	0.8209	0.8848	0.8715	0.8925	0.8530	0.8755	0.8496	0.8958	0.8743
	Sampling	0.9461	0.9482	0.9474	0.8884	0.9352	0.9565	0.9576	0.9683	0.9569	0.9629	0.9579	0.9628
Hit number		1	4	6	1	11	2	10	21	3	17	7	22

Back to article page