We will carry out the quantified adaptive fusion of the ``card of the favorable zones'' and the spectral bands to verify the efficiency of the fusion by priority. All the sources are thus put at the same level, without any preference.

Table 39 shows the rates of classification obtained and the resultant image. The rates of classification are comparable with those of fusion by priority. The result image presents an aspect more disturbed however than that obtained with fusion by priority. In effect, the ``card of the areas favourable to each class'', when it is placed on the same level than the spectral bands, is rather lost among spectral information. It becomes then more difficult to decide according to its opinion because no difference is made between it and the spectral bands.

For fusion by priority, in the event of indecision on behalf of the spectral bands (several classes suggested with close degrees of possibility), preference is given to the opinion expressed by the most reliable source. On the other hand, when the spectral bands agree clearly to support a particular class i, this one is generally retained, even if it is in contradiction with the information provided by the ``card of the favourable zones''.

Finally, in the event of confusion, the most reliable source prevails. And in the event of agreement of a majority of sources, those sources can bring the final decision, even if this decision is different from that preferred by the most reliable source.

Table 39: Quantified adaptive fusion of the 4 spectral bands and of the source ``card of the areas favourable to each source''.

Rates of classification obtained
Class	Number of pixels   correctly classified (A)	Number of pixels in samples      (B)	Rate of pixels     correctly classified (A/B)
1	442	459	96.30%
2	421	459	91.72%
3	258	306	84.31%
4	368	391	94.12%
5	399	459	86.93%
6	437	459	95.21%
7	448	459	97.60%
8	447	459	97.39%
9	428	459	93.25%
TOTAL	3648	3910	93.30%

Table 40: Inter-classes confusion matrix relative to the ``card of the areas favourable to each source''.

	Class observed
Class expected	1	2	3	4	5	6	7	8	9	Unclassified
1	99.78	-	-	-	-	0.22	-	-	-	-
2	3.92	96.08	-	-	-	-	-	-	-	-
3	-	-	96.73	3.27	-	-	-	-	-	-
4	-	-	0.51	94.12	5.37	-	-	-	-	-
5	-	-	-	20.48	79.52	-	-	-	-	-
6	8.50	-	-	-	-	91.50	-	-	-	-
7	-	-	0.65	0.22	-	-	97.60	1.53	-	-
8	-	-	-	-	-	-	-	100.00	-	-
9	-	-	-	-	-	-	-	-	100.00	-

Table 41: Inter-classes confusion matrix (percentages) relative to classification after fusion by priority of groups K₁ (``card of the areas favourables to each class'') and K₂ (the 4 spectral bands MSS4 to MSS7).

	Class observed
Class expected	1	2	3	4	5	6	7	8	9	Unclassified
1	93.68	-	1.53	3.05	-	1.74	-	-	-	-
2	3.92	88.45	0.22	5.66	1.74	-	-	-	-	-
3	-	-	96.41	2.61	0.98	-	-	-	-	-
4	-	-	1.53	94.12	4.35	-	-	-	-	-
5	-	-	7.19	9.59	82.35	0.87	-	-	-	-
6	1.74	-	-	4.36	-	93.68	0.22	-	-	-
7	-	-	-	0.65	-	-	95.86	3.27	0.22	-
8	-	-	-	-	-	-	1.09	97.60	1.31	-
9	-	-	-	0.22	-	-	3.27	-	96.51	-

Table 42: Inter-classes confusion matrix (percentages) relative to classification after adaptive quantified fusion of MSS4 to MSS7 spectral bands and of the source ``card of the areas favourables to each class''.

	Class observed
Class expected	1	2	3	4	5	6	7	8	9	Unclassified
1	96.30	0.44	0.22	2.18	-	0.87	-	-	-	-
2	3.70	91.72	-	2.40	2.18	-	-	-	-	-
3	-	-	84.31	4.25	8.17	-	3.27	-	-	-
4	-	-	0.51	94.12	5.37	-	-	-	-	-
5	-	-	1.53	11.11	86.93	0.22	0.22	-	-	-
6	2.83	0.44	-	0.87	0.22	95.21	0.44	-	-	-
7	-	-	0.65	0.22	-	-	97.60	1.53	-	-
8	-	-	-	-	-	-	2.61	97.39	-	-
9	-	-	-	-	-	-	6.75	-	93.25	-