Non-attention region first initialisation of k-means clustering for saliency detection

• Published in: Electronics letters Vol. 49; no. 22; pp. 1384 - 1386
• Main Authors: Chen, Jiazhong, Cao, Hua, Ju, Zengwei, Qin, Leihua, Su, Shuguang
• Format: Journal Article
• Language: English
• Published: Stevenage The Institution of Engineering and Technology 24.10.2013; Institution of Engineering and Technology; John Wiley & Sons, Inc
• Subjects: Applied sciences; Artificial intelligence; Cluster analysis; Clustering; Color; Colour; colour contrast; Computer science; control theory; systems; cut‐off low‐frequency; Exact sciences and technology; Image and vision processing and display technology; image colour analysis; k‐means clustering; multiwavelet reconstruction; NARFI; nonattention region first initialisation; object detection; pattern clustering; Pattern recognition. Digital image processing. Computational geometry; Reconstruction; saliency detection; saliency map generation; Seeds; k-means clustering; saliency detection; pattern clustering; nonattention region first initialisation; colour contrast; object detection; image colour analysis; NARFI; cut-off low-frequency; multiwavelet reconstruction; saliency map generation; Image processing; Wavelet transformation; Contrast; Color; Pattern clustering
• ISSN: 0013-5194; 1350-911X; 1350-911X
• DOI: 10.1049/el.2013.2181

According to the nature of saliency map generation with colour contrast, a non-attention region first initialisation (NARFI) k-means clustering for saliency detection is proposed. The NAR is obtained by multiwavelet reconstruction based on the cutoff low-frequency. The initial seeds of the k-means are chosen from the NAR. This way, the NAR is clustered in a fine manner, whereas the attention region is clustered in a coarse manner. As a result, the saliency values of the attention region with the NARFI k-means clustering are more conspicuous than those with the k-means++ clustering.