今天，試一試----->基於引導濾波和均值濾波的多聚焦圖像融合---->實現。實現異源圖像的融合。
- Today, try -> multi-focus image fusion based on guided filtering and mean filtering -> implementation.Realize the fusion of different images.
看看結果：
- result：
解釋：
對於代碼的實現，其實底層就是一個引導濾波器，基於引導濾波和均值濾波的多聚焦圖像融合。看效果還是很不錯的。但是是否適用於我們自己的數據集，還是得通過多種評價指標來判定。
圖像融合的評價指標我進行了相應的總結：
-
Explanation:
For code implementation, in fact, the bottom layer is a guide filter, based on the guide filter and mean filter multi-focus image fusion.It looks pretty good.However, whether it is applicable to our own data set depends on a variety of evaluation indicators.
The evaluation indexes of image fusion are summarized as follows:
圖像質量評價
圖像算法評估
圖像算法評估 = 定性（主觀，觀察） + 定量（客觀，特徵值） + 算法時間
定性：主要是觀察+分析
定量：主要是各參數指標，又分爲 參考質量+非參考質量
參考質量：處理後的圖和原圖之間的相關質量度量，比如對比度提升，輪廓復原率，過飽和率，結構相似度，
PSN, SSIM, RMSE and UIQ
非參考質量：圖像本身的一些指標，
e, σ, r and CNR
-
Image quality evaluation
Image algorithm evaluation
Image algorithm evaluation = qualitative (subjective, observation) + quantitative (objective, eigenvalue) + algorithm time
Qualitative: basically be observation + analysis
Quantitative: mainly refers to each parameter index, which is also divided into reference quality + non-reference quality
Reference quality: Relative quality measures between the processed image and the original image, such as contrast enhancement, contour recovery rate, supersaturation rate, structural similarity,
PSN, SSIM, RMSE and UIQ
Non-reference quality: some indicator of the image itself,
E,, R and CNR
圖像質量主觀評價
絕對評價
相對評價
-
Subjective evaluation of image quality
Absolute evaluation
Relative evaluation
圖像質量客觀評價
- Objective evaluation of image quality
我已經實現了14中圖像質量的評價算法，到時候根據自己的需求進行選擇合適的評價算法。
-
I have implemented 14 image quality evaluation algorithms, and then I will choose the appropriate evaluation algorithm according to my own needs.
I hope I can help you,If you have any questions, please comment on this blog or send me a private message. I will reply in my free time.