search

jacob6 / ENIQA

Entropy-based No-reference Image Quality Assessment
42 stars 5 forks source link

Why is the ENIQA score negative? #2

Open angleboy8 opened 2 years ago

angleboy8 commented 2 years ago

When I test an enhanced image, the ENIQA score is negative. The range of the output score is expected to be [0,1] and a lower score represents higher quality. For this problem，I have some confused. If possible, could you give a reasonable explanation. I appreciate you very much.

jacob6 commented 2 years ago 
font{
    line-height: 1.6;
}
ul,ol{
    padding-left: 20px;
    list-style-position: inside;
}

When we deal with IQA tasks, we generally assumed that the image to be tested is degraded, not enhanced

On 7/13/2021 ***@***.***> wrote：

When I test an enhanced image, the ENIQA score is negative. The range of the output score is expected to be [0,1] and a lower score represents higher quality. For this problem，I have some confused. If possible, could you give a reasonable explanation. I appreciate you very much.

—You are receiving this because you are subscribed to this thread.Reply to this email directly, view it on GitHub, or unsubscribe.

angleboy8 commented 2 years ago

I apreciate your reply and help. In other words, ENIQA is not used to assess the image quality of restored images and is only applied to degraded images, such as noisy or blurred images. Please correct me if my understanding is wrong. Thank you again! Best wishes.

---Original--- From: "Guangyi @.> Date: Thu, Jul 15, 2021 09:30 AM To: @.>; Cc: @.**@.>; Subject: Re: [jacob6/ENIQA] Why is the ENIQA score negative? (#2)

 font{ 
     line-height: 1.6; 
 } 
 ul,ol{ 
     padding-left: 20px; 
     list-style-position: inside; 
 } 

 When we deal with IQA tasks, we generally assumed that the image to be tested is degraded, not enhanced

 On 7/13/2021 ***@***.***> wrote：

When I test an enhanced image, the ENIQA score is negative. The range of the output score is expected to be [0,1] and a lower score represents higher quality. For this problem，I have some confused. If possible, could you give a reasonable explanation. I appreciate you very much.

—You are receiving this because you are subscribed to this thread.Reply to this email directly, view it on GitHub, or unsubscribe. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

angleboy8 commented 2 years ago

font{ line-height: 1.6; } ul,ol{ padding-left: 20px; list-style-position: inside; } When we deal with IQA tasks, we generally assumed that the image to be tested is degraded, not enhanced On 7/13/2021 @.***> wrote： When I test an enhanced image, the ENIQA score is negative. The range of the output score is expected to be [0,1] and a lower score represents higher quality. For this problem，I have some confused. If possible, could you give a reasonable explanation. I appreciate you very much. —You are receiving this because you are subscribed to this thread.Reply to this email directly, view it on GitHub, or unsubscribe.

I appreciate your reply and help. In other words, ENIQA is not used to assess the image quality of restored images and is only applied to degraded images, such as noisy or blurred images. Please correct me if my understanding is wrong.

jacob6 commented 2 years ago 
font{
    line-height: 1.6;
}
ul,ol{
    padding-left: 20px;
    list-style-position: inside;
}

Your understanding is correct. So far, there are few studies on the quality assessment of enhanced images, so you can try it.

On 7/15/2021 ***@***.***> wrote：

font{ line-height: 1.6; } ul,ol{ padding-left: 20px; list-style-position: inside; } When we deal with IQA tasks, we generally assumed that the image to be tested is degraded, not enhanced On 7/13/2021 @.***> wrote： When I test an enhanced image, the ENIQA score is negative. The range of the output score is expected to be [0,1] and a lower score represents higher quality. For this problem，I have some confused. If possible, could you give a reasonable explanation. I appreciate you very much. —You are receiving this because you are subscribed to this thread.Reply to this email directly, view it on GitHub, or unsubscribe.

I appreciate your reply and help. In other words, ENIQA is not used to assess the image quality of restored images and is only applied to degraded images, such as noisy or blurred images.

Please correct me if my understanding is wrong.

—You are receiving this because you commented.Reply to this email directly, view it on GitHub, or unsubscribe.

angleboy8 commented 2 years ago

font{ line-height: 1.6; } ul,ol{ padding-left: 20px; list-style-position: inside; } Your understanding is correct. So far, there are few studies on the quality assessment of enhanced images, so you can try it. On 7/15/2021 @.> wrote： font{ line-height: 1.6; } ul,ol{ padding-left: 20px; list-style-position: inside; } When we deal with IQA tasks, we generally assumed that the image to be tested is degraded, not enhanced On 7/13/2021 @.> wrote： When I test an enhanced image, the ENIQA score is negative. The range of the output score is expected to be [0,1] and a lower score represents higher quality. For this problem，I have some confused. If possible, could you give a reasonable explanation. I appreciate you very much. —You are receiving this because you are subscribed to this thread.Reply to this email directly, view it on GitHub, or unsubscribe. I appreciate your reply and help. In other words, ENIQA is not used to assess the image quality of restored images and is only applied to degraded images, such as noisy or blurred images. Please correct me if my understanding is wrong. —You are receiving this because you commented.Reply to this email directly, view it on GitHub, or unsubscribe.

I don't know blind/referenceless/non-reference image quality assessment (IQA) very well. I test extensive IQA models but most algorithms is not applied to restored results. Thanks for your advice. I wish that you will provide more excellent research.