(Eng.) Weighted Color Structure Code (WCSC)

Image segmentation is a fundamental step in several image processing tasks. It is a process where an image is divided into its constituent regions guided by a similarity criterion. One very interesting image segmentation method is the Color Structure Code (CSC), which combines simultaneously split-and-merge and region-growing techniques. In this paper a segmentation approach based on the CSC method, Weighted Color Structure Code (WCSC), is proposed. This method is guided by a non-linear discrimination function, where the user-inference is captured by the Polynomial Mahalanobis, prioritizing, during the merging process, the regions with higher similarity to the user selected pattern. The WCSC has color distribution pattern-oriented characteristic, showing better coherence among the segments with higher similarity to the selected pattern. A qualitative evaluation and parametric paired analysis were performed to compare CSC, WCSC and other segmentation methods results, using images from Berkeley benchmark. The results from these comparison indicate an improvement on the segmentation result obtained by the WCSC.

First Experiment

The perform experiment was the comparison, first visual than quantitative of the results for both methods.

Segmentation Results obtained using 60 images from Berkeley Dataset:

Table below shows the values of Rand index for each of the 60 images, used for validation and quantitative analysis:

Table comparing the Rand index results between CSC and WCSC for all the 60 selected images. The column image is the reference number of the image on the Berkeley dataset.

Statistical comparison between the values of Rand index for the two approaches:

Comparison of CSC and WCSC performance. (a) box-plot comparing CSC and WCSC Rand index dissimilarity to GT results. (b) paired comparison image sequence related to GT.

Second Experiment

The second experiment uses the results obtained previously for both methods (CSC and WCSC) and compares with other segmentation algorithms (Edge Detection and Image Segmentation (EDISON), Mumford-Shah (MS), Watershed (WS), JSEG, Recursive Hierarchical Segmentation (RHSEG)). Figure below shows the comparison between segmentation results for all the algorithms used and the graph result from Rand index value for each method.

Comparison between segmentation methods on image 368068. (A) original image, (B)
Ground Truth, (C) WCSC, (D) CSC, (E) Edison, (F) Munford-Shah, (G) RHSEG, (H) JSEG and
(I) Watershed.

Graph showing the Rand index score for the image 368068 for each method tested.

To summarize the comparison between methods, we show in Figure below the boxplot of the Rand index obtained for 16 images with each tested method and the values of Rand index used to build the boxplot. This plot shows that WCSC has the lowest mean value, indicating that among the 16 images, the proposed approach, according to the validation method, obtained an improvement on the segmentation results.

Graph showing the boxplot of Rand index for 16 images for each method tested.