COMPARISON OF ALGORITHMS USED IN UNDERWATER IMAGE ENHANCEMENT

Birkan Büyükarıkan

Makale detayları

COMPARISON OF ALGORITHMS USED IN UNDERWATER IMAGE ENHANCEMENT

Yazarlar:

Birkan Büyükarıkan

Yayıncı:

Isparta University of Applied Sciences

Ana Başlıklar:

Yapay Görme | Aydınlatma

Dergi:

Yalvaç Akademi Dergisi Vol.9 Issue 1, pp.33-45

Belge Tipi:

Makale

Yayın Tarihi:

3/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001313459

pISSN:

2548-0820

Anahtar kelimeler:

Özet:

 The exploration of the underwater world has become an intriguing research subject in recent years. However, factors such as blurriness, color distribution, and contrast in the underwater environment lead to distortions such as noise and loss of detail in underwater images. This situation increases the challenges encountered in underwater image analysis applications. Image enhancement algorithms can be employed to overcome these problems. The aim of this study is to improve underwater images using existing image enhancement algorithms. Single Scale Retinex SSR, Multi Scale Retinex MSR, Multi Scale Retinex with Color Restoration MSRCR, Dehazing Histogram Distribution Prior DHDP, and Multi-Scale Correlated Wavelet MSCW are among the enhancement algorithms used in this study. This study was applied to images selected from a data set containing real-world data. Evaluations with both full-reference and no-reference metrics were conducted to demonstrate the enhancement algorithms performance. According to the results obtained from the evaluation metrics of the selected images, images enhanced with the MSRCR algorithm generally achieved better results on average. The average scores for the MSRCR algorithm in Peak Signal-to-Noise Ratio, Structural Similarity Index, Blindreferenceless Image Spatial Quality Evaluator, Naturalness Image Quality Evaluator, Perception based Image Quality Evaluator, Underwater Image Quality Measure, and Underwater Color Image Quality Evaluation, underwater image quality criterion, and underwater colored image quality evaluation criteria are 15.6454, 0.4516, 22.6035, 6.4106, 34.7032, 1.7344, and 7.776, respectively. Experimental results demonstrate the effectiveness of image enhancement methods in reducing distortions in underwater images.

 The exploration of the underwater world has become an intriguing research subject in recent years. However, factors such as blurriness, color distribution, and contrast in the underwater environment lead to distortions such as noise and loss of detail in underwater images. This situation increases the challenges encountered in underwater image analysis applications. Image enhancement algorithms can be employed to overcome these problems. The aim of this study is to improve underwater images using existing image enhancement algorithms. Single Scale Retinex SSR, Multi Scale Retinex MSR, Multi Scale Retinex with Color Restoration MSRCR, Dehazing Histogram Distribution Prior DHDP, and Multi-Scale Correlated Wavelet MSCW are among the enhancement algorithms used in this study. This study was applied to images selected from a data set containing real-world data. Evaluations with both full-reference and no-reference metrics were conducted to demonstrate the enhancement algorithms performance. According to the results obtained from the evaluation metrics of the selected images, images enhanced with the MSRCR algorithm generally achieved better results on average. The average scores for the MSRCR algorithm in Peak Signal-to-Noise Ratio, Structural Similarity Index, Blindreferenceless Image Spatial Quality Evaluator, Naturalness Image Quality Evaluator, Perception based Image Quality Evaluator, Underwater Image Quality Measure, and Underwater Color Image Quality Evaluation, underwater image quality criterion, and underwater colored image quality evaluation criteria are 15.6454, 0.4516, 22.6035, 6.4106, 34.7032, 1.7344, and 7.776, respectively. Experimental results demonstrate the effectiveness of image enhancement methods in reducing distortions in underwater images.

 The exploration of the underwater world has become an intriguing research subject in recent years. However, factors such as blurriness, color distribution, and contrast in the underwater environment lead to distortions such as noise and loss of detail in underwater images. This situation increases the challenges encountered in underwater image analysis applications. Image enhancement algorithms can be employed to overcome these problems. The aim of this study is to improve underwater images using existing image enhancement algorithms. Single Scale Retinex SSR, Multi Scale Retinex MSR, Multi Scale Retinex with Color Restoration MSRCR, Dehazing Histogram Distribution Prior DHDP, and Multi-Scale Correlated Wavelet MSCW are among the enhancement algorithms used in this study. This study was applied to images selected from a data set containing real-world data. Evaluations with both full-reference and no-reference metrics were conducted to demonstrate the enhancement algorithms performance. According to the results obtained from the evaluation metrics of the selected images, images enhanced with the MSRCR algorithm generally achieved better results on average. The average scores for the MSRCR algorithm in Peak Signal-to-Noise Ratio, Structural Similarity Index, Blindreferenceless Image Spatial Quality Evaluator, Naturalness Image Quality Evaluator, Perception based Image Quality Evaluator, Underwater Image Quality Measure, and Underwater Color Image Quality Evaluation, underwater image quality criterion, and underwater colored image quality evaluation criteria are 15.6454, 0.4516, 22.6035, 6.4106, 34.7032, 1.7344, and 7.776, respectively. Experimental results demonstrate the effectiveness of image enhancement methods in reducing distortions in underwater images.

 Su altının keşfi, son yıllarda ilginç bir araştırma konusu olmuştur. Ancak su altı ortamındaki bulanıklık, renk dağılımı ve kontrast gibi etkenler, su altı görüntülerinde gürültü ve ayrıntı kaybı gibi bozulmalara neden olur. Bu durum, su altı görüntü analizi uygulamalarında karşılaşılan zorlukları arttırır. Bu sorunların üstesinden gelebilmek için görüntü iyileştirme algoritmaları kullanılabilir. Bu çalışmanın amacı, mevcut görüntü iyileştirme algoritmaları yardımıyla su altı görüntülerinin geliştirilmesidir. Çalışmada mevcut iyileştirme algoritmalarından; tek-ölçekli retineks SSR, çok ölçekli retineks MSR, renk düzeltmeli çok ölçekli retineks MSRCR, öncelikli histogram dağıtımı DHDP ve çok ölçekli ilişkili dalgacık MSCW kullanılmıştır. Çalışma, gerçek dünya verilerini içeren bir veri setinden seçilen görüntülere uygulanmıştır. Kullanılan iyileştirme algoritmalarının performanslarını göstermek için tam referanslı ve referanssız ölçütlerle değerlendirme yapılmıştır. Seçilen görüntülerin değerlendirme ölçütlerinden elde edilen sonuçlara göre MSRCR algoritmasıyla iyileştirilen görüntülerde ortalama olarak daha iyi sonuçlar elde edilmiştir. MSRCR algoritmasının tepe sinyal-gürültü oranı, yapısal benzerlik indeksi, körreferanssız görüntü uzamsal kalite değerlendiricisi, doğallık görüntü kalitesi değerlendiricisi, algı tabanlı görüntü kalitesi değerlendiricisi, su altı görüntü kalitesi ölçütü ve su altı renkli görüntü kalitesi değerlendirme ölçütlerinin ortalama puanları sırasıyla 15,6454, 0,4516, 22,6035, 6,4106, 34,7032, 1,7344 ve 7,776dır. Deneysel sonuçlar, su altı görüntülerinde bozulmaları azaltmak için görüntü iyileştirme yöntemlerinin kullanılabilirliğini göstermektedir.

Makale detayları

COMPARISON OF ALGORITHMS USED IN UNDERWATER IMAGE ENHANCEMENT

Yazarlar:

Birkan Büyükarıkan

Yayıncı:

Isparta University of Applied Sciences

Ana Başlıklar:

Yapay Görme | Aydınlatma

Dergi:

Yalvaç Akademi Dergisi Vol.9 Issue 1, pp.33-45

Belge Tipi:

Makale

Yayın Tarihi:

3/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001313459

pISSN:

2548-0820

Underwater Images

Image Enhancement

Image QualitySu Altı Görüntüler

Görüntü İyileştirme

Görüntü Kalitesi

Sayı

Makale detayları

COMPARISON OF ALGORITHMS USED IN UNDERWATER IMAGE ENHANCEMENT

Yazarlar: Birkan Büyükarıkan

Yayıncı: Isparta University of Applied Sciences

Ana Başlıklar: Yapay Görme | Aydınlatma

Dergi: Yalvaç Akademi Dergisi Vol.9 Issue 1, pp.33-45

Belge Tipi: Makale

Yayın Tarihi: 3/2024

Belge Dili: Türkçe

Document ID (Org No.): 2001313459

pISSN: 2548-0820

Underwater Images

Image Enhancement

Image QualitySu Altı Görüntüler

Görüntü İyileştirme

Görüntü Kalitesi

Sayı

Yazarlar:

Birkan Büyükarıkan

Yayıncı:

Isparta University of Applied Sciences

Ana Başlıklar:

Yapay Görme | Aydınlatma

Dergi:

Yalvaç Akademi Dergisi Vol.9 Issue 1, pp.33-45

Belge Tipi:

Makale

Yayın Tarihi:

3/2024

Belge Dili:

Türkçe

Document ID (Org No.):

2001313459

pISSN:

2548-0820