Focus Measurement in Color Space for Shape From Focus Systems

Abstract

Shape from Focus (SFF) has been studied extensively in computer vision for 3D shape and depth recovery. The first stage in SFF methods is to compute the focus value of every pixel by converting the colored images into gray scale and then apply the focus measure operator. Converting colored values in the images into gray scale values may lead to imprecise mapping of pixels with different colored values onto the same gray scale value, this affects the overall accuracy of the system. In a colored image, the focused pixels maintain a considerable color difference from their neighboring pixels as compared to the defocused ones, which are blended into their neighborhood. This article presents an alternative method to measure the degree of focus by directly processing colored images. The color differences of the neighbor pixels with respect to the central pixel are obtained and summed together, this is followed by calculating their spread. The sum and the spread are combined to measure the degree of focus of the pixel in consideration. The proposed focus measure is then used for shape recovery of various simulated and real objects and is compared with previous techniques. The comparison results show the proposed method has the highest correlation and smallest RMSE values confirming the effectiveness of using color images for shape recovery.

3D reconstruction of gastrointestinal regions using shape-from-focus

“…After image acquisition, next step is to measure the focus value of each pixel in the image stack. A color focus measure is applied on RGB images to compute the focus value of each pixel [12].…”

“…It is then followed by calculating their spread. The sum and the spread can be combined together as [12],…”

“…These FM's consider gray scale images for depth recovery. Recently, color FM are also proposed which uses RGB information in the image to evaluate the quality of a pixel [11,12].…”

Directional Ring Difference Filter for Robust Shape-from-Focus

“…However, the results also indicate that increasing the sizes of the filters for both RDF and DRDF leads to a greater deviation from the ground truth and, hence, results in a loss of details. Next, the proposed method was compared to seven focus computation methods, i.e., gray-level variance (GLV) [7], modulus of the gradient of the color channel (MCG) [13], modified-Laplacian (ML) [3], sum and spread focus measure (FMSS) [23], reduced Tenengrad (RT) [14], multi-scale-morphological focus measure (MSM) [24], and ring difference filter (RDF) [25]. For visual comparisons, we constructed depth maps of synthetic datasets, Antinous, Cotton, and Pens, using different methods, as shown in Figure 5.…”

“…Additionally, there are some other FMs that do not fall in the previously mentioned categories but produce exceptional results. For example, the sum and spread focus measure (FMSS) [23] that calculates an image's sharpness value by using basic vector operations and incorporating information from different color channels. A multi-scale morphological focus measure (MSM) [24] that uses morphological operations i.e., dilation and erosion to obtain sharpness values and integrates them on different scales.…”