Estimation of weighting distribution using fuzzy memberships and wavelet transformation with PSO optimization in satellite image enhancement

Figures & data

Figure 1. The 2D-DWT decomposition.

Figure 2. Block diagram of the proposed algorithm.

Table

Step 1: Read the low-resolution and low-contrast multi-spectral satellite image A.
Step 2: Histogram equalization is performed on Image A that yields an image B.
Step 3: DWT decomposition is performed on satellite images A and on histogram equalized image B, respectively. We obtain the corresponding coefficients Low–Low, Low–High, High–Low, High–High of low resolution satellite image A are $\mathrm{\Phi }(LL,A),\mathrm{\Phi }(LH,A),\mathrm{\Phi }(HL,A),\mathrm{\Phi }(HH,A)$, respectively, and coefficients Low–Low, Low–High, High–Low, High–High of histogram equalized satellite image B are $\mathrm{\Phi }(LL,B),\mathrm{\Phi }(LH,B),\mathrm{\Phi }(HL,B),\mathrm{\Phi }(HH,B)$, respectively
Step 4: The high-frequency subband for each color satellite images $\mathrm{\Phi }(HH,A)$ and $\mathrm{\Phi }(HH,B)$ are bifurcated into three channels $R_A,G_A,B_A$, respectively, and $R_B,G_B,B_B$, respectively.
Step 5: Optimize parameters r, t, and s using PSO and forms the fuzzy edge map for the resultant HH subbands. This band is formed by compute the objective function in given equation no 5. The maximum value of objective function preserves the edge strength present in the satellite image.
Step 6: The new color fuzzy edge map is defined on the basis of fusion using equation no 8 of three resulted in color channel of HH subband detail information $w_A$ and $w_B$.
Step 7: The detail information coefficient is combined using fuzzy membership function
equation 1 and equation 3.
Step 8: Apply the inverse discrete wavelet transform (IDWT) sequentially in order to modify
the higher domain coefficients using equation 11 with other all subbands.
Step 9: Finally, the enhanced satellite image has resulted

Table 1. Comparison of luminance, entropy, average contrast, CAF values using proposed approach, DWT- and DCT-based methods over different images

Images		Luminance $\overline{L}$	Entropy H	Average contrast $\overline{C}$	CAF
1	Test image	79.9986	6.4551	65.6653	18.3753
	Proposed approach	98.4167	7.6711	194.7099	28.6551
	DWT (Demirel & Anbarjafari, 2011)	128.3178	7.1251	93.4568	22.1536
	DCT (Bhandari et al., 2011)	89.4299	6.6564	73.0664	19.4610
2	Test image	65.4602	7.0769	99.3299	22.3417
	Proposed approach	112.9181	7.4293	201.7765	28.0005
	DWT (Demirel & Anbarjafari, 2011)	116.2298	7.4915	146.0859	26.0447
	DCT (Bhandari et al., 2011)	85.2260	7.3169	127.5119	24.5875
3	Test image	71.6810	6.6087	54.6698	17.9702
	Proposed approach	137.1429	7.3670	216.1301	28.2469
	DWT (Demirel & Anbarjafari, 2011)	172.8170	6.9092	95.3200	21.5887
	DCT (Bhandari et al., 2011)	101.3466	7.1918	81.3107	21.5960
4	Test image	39.8276	5.8954	56.0258	16.1292
	Proposed approach	97.0773	7.6711	134.7808	26.1376
	DWT (Demirel & Anbarjafari, 2011)	110.2186	6.7147	140.0949	23.1010
	DCT (Bhandari et al., 2011)	85.5918	6.7959	144.7823	23.5737
5	Test image	56.6508	6.5098	75.0127	19.1581
	Proposed approach	109.4173	7.7799	138.2868	26.6788
	DWT (Demirel & Anbarjafari, 2011)	131.7408	6.1716	134.5537	21.0195
	DCT (Bhandari et al., 2011)	104.7732	7.0262	146.8237	24.4580
6	Test image	107.9119	6.3510	68.8707	18.2957
	Proposed approach	119.0543	7.8674	218.2253	30.2382
	DWT (Demirel & Anbarjafari, 2011)	174.3260	6.8506	90.1928	21.1115
	DCT (Bhandari et al., 2011)	118.0860	6.6610	82.7390	20.0893
7	Test image	97.8874	7.0313	58.4979	19.4455
	Proposed approach	111.2269	7.9427	111.4781	25.8086
	DWT (Demirel & Anbarjafari, 2011)	152.1536	7.2286	72.3521	21.0824
	DCT (Bhandari et al., 2011)	89.4299	6.6564	73.0664	19.4610

Table 2. Comparison of EME values for proposed approach, DWT- and DCT-based methods

Test image	Proposed approach	DWT based method (Demirel & Anbarjafari, 2011)	DCT based method (Bhandari et al., 2011)
Image 1	14.1886	3.8198	4.6112
Image 2	11.9619	7.9285	9.4718
Image 3	12.7564	5.6743	3.4208
Image 4	8.5995	7.8576	5.6761
Image 5	9.2419	7.3345	3.4930
Image 6	7.7436	4.8682	7.2640
Image 7	11.0726	3.0337	4.6112

Figure 3. Resulted images.

Demirel, H., & Anbarjafari, G. (2011). Discrete wavelet transform-based satellite image resolution enhancement. IEEE Transactions on Geoscience and Remote Sensing, 49(6), 1997–2004.

 Web of Science ®Google Scholar

Bhandari, A. K., Kumar, A., & Padhy, P. K. (2011). Enhancement of low contrast satellite images using discrete cosine transform and singular value decomposition. World Academy of Science, Engineering and Technology, 79, 35–41.

 Google Scholar