ŶŮŢŨŢŻŪůŦġŪŴŴŶŦġljĸ ĵĵ 儎▊⅙是澳門大學科ᢶ學院ᮮ學系副教授，研究乎ตऻᤢ 㼽࠼᷆ 、 ݹ ㍖࠼᷆ 、ڻ 㩿࠼᷆ 、ᱸ乱࠼᷆ ，ԛ及䁀㲕㠽ൌ܅㲋⨼。ⴤ ࢃ主㾷研究方向是 ݹ ㍖࠼᷆ 及ެ 在䁀㲕ૂ ൌ܅㲋⨼中的⭞。 最䘇⦨ᗍҼ 2016 201 年ᓜ科ᢶ學院研究 ݠ ⿶⦄，ќ在ੂ 行䂋ሟ 的國䳑期刊р發㺞Ҽ 0 ཐㇽ䄌文。 The author is an associate professor in the Department of Mathematics, Faculty of Science and Technology, University of Macau (UM). Her research interests include complex analysis, Quarternion, Fourier analysis, time frequency analysis, and signal and image processing. Currently she mainly focuses on Quarternion analysis and its applications in signal and image processing. She recently received the FST Research Excellence Award 2016/2017 and has published more than 50 papers in peer-reviewed international journals. ཇ的Sobel ᙭ޯ ᐡ，Canny ႻᇖЮᕭเᐡ ۍ գዃ 的ᕕӡ，֭ ࣏ഺ ڱ ᕭเᐡҪւዋۖ ؊ൊ的ࢧነᢏ м，પ־࣏ᛞ֒ 、ࠪ ࢘ຈ的ࢧነᢏм。ഺ ڱ પወ ࢝ጀᕢႵऎ᎑ᛔ的ඞ۶ᘳ的ࠫ ਿ，഼Ⴥࠪ ࢘，ᛞ֒ ᢏм ڭ Й ږ ѽ د ፡ᕭเஇՀ。ऎϞ ـ Հ Գໞเ፡ᕭเᆊᇖ ا，ޱ েु ٿ ೡ࢙ӲϞϘ ڀ ԕ 的ह֜ ᕭเᐡ，ᐯ೫ࢨዹᗑԪᛞ֒ 、ࠪ ࢘、࢘ഺ ڱ પወ ڽ இ፡ᕭเ，Ҩஇह֜ ፡ᕭเᐡ，ᚎᇍ QDPC۶QDPA。ᄦljġ࣏ంᇨϯ ۖر 的Ϭเႏ的 ᄦᄊ， ݗ Є、 ࠅ ۶ ڄ ᡙ， ا েѧႵԧᇌ፡ᕭเ ᐡ的 ࠑ ġĩᄦijĪ。 ంᄦijКҞѽԪನԳఀ ҍຕᎢ，ᅆ ݺ ཇᕭเᐡ Sobel，Canny ا， েȞQDOC、QDPAў ޱ ȟգႵ Հ的 ࠑ 。Ԑ， ݗ ᄦᄊ的ຕ ，ށࠑލ ݺ ह֜ 的ў ޱ ĩDPC、MDPC、QDPC、QDPAȟҞѽࠧ ௱ Հ的 ࠑ ݗ צ 的ြ。ġह֜ 的ᆊᇖ ޱ ϵҞѽ ࠅ ం०К෩ۤ 。ġSobel ў ޱ ϵҞѽం०К෩ۤ ࠅ է，փCannyў ޱ Йକ ࠅ ሃ०ж༟。 ێ ը， ᅆ ڄݺ ᡙᄦᄊ，QDPAў ۖఀޱ Йᓾ的፡ᕭเຕ ，ލ ᓾሱႵь， ࠑ Ⴕ ێ ҁᆊᇖ ޱ ನྐ 。 for short. Figure 1 contains three images found on the internet, of a house, a "ower, and a liver, respectively. We then compared the performance of di#erent phase edge detectors. As shown in !gure 2, QDPC and QDPA performed better than Sobel and Canny. Firstly, the image of the house shows that phase-based methods (DPC, MDPC, QDPC, QDPA) can perform very well in presenting the details of the chimney of the house. Both phase-based methods and Sobel can extract the "ower from the background. But Canny cannot separate the "ower from the background. Secondly, for the liver image, QDPA performed better than the other methods, with fewer errors and better results. ൌ2 )LJXUH2 ⭧ᐜ㠩是࠼࡛֒ ⭞Sobel、Canny、DPC、MDPC、QDPC、QDPA⃘ ③ಞ的㎆᷒ 。 (From left to right) results from Sobel，Canny，DPC，MDPC，QDPC，and QDPA. Neither Canny nor Sobel can detect the shaded regions of the liver. To suppress noise, Canny edge detector !rst smoothed out the image. But this process can cause the loss of edge points. So, Canny treated the shaded regions as noise and deleted them. Moreover, there is no signi!cant di#erence between the shaded regions and their surrounding regions in terms of colour. Sobel failed to detect the shaded regions. $e DPC, MDPC, QDPA and QDPC are all based on the phase methods. $ese methods include the properties of an image. We can see that these methods can detect the shaded regions of the liver clearly. QDPA not only can detect the smooth areas, but can also accurately and clearly detect the shaded regions of the liver. ŅŦŵŦŤŵŪůŨġŔũŢťŦťġœŦŨŪŰůŴġ ŰŧġŵũŦġōŪŷŦų
RkJQdWJsaXNoZXIy MTQ1NDU2Ng==