An algorithm for mobile vision-based localization of skewed nutrition labels that maximizes specificity - Emerging Trends in Image Processing, Computer Vision, and Pattern Recognition - page 233

recall and precision may not mater much because of the fast rate at which input images are
processed on target devices, but there is definitely room for improvement.

Table 1
NL Localization Results

PR TR CR PR SP ACC

0.7632 0.422 0.3580 0.1475 1.0 0.5916

4.3 Limitations
The majority of false negative matches were caused by blurry images. Blurry images are the
result of poor camera focus and instability. Both the CED and dilate-erode corner detector re-
quire rapid and contrasting changes to identify key points and lines of interest. These points
and lines are meant to correspond directly with text and NL borders. These useful data can-
not be retrieved from blurry images, which results in run-time detection failures. The only re-
course to deal with blurry inputs is improved camera focus and stability, both of which are
outside the scope of this algorithm, because it is a hardware problem. It is likely to work bet-
ter in later models of smartphones. The current implementation on the Android platform at-
tempts to force focus at the image center but this ability to request camera focus is not present
in older Android versions. Over time, as device cameras improve and more devices run newer
versions of Android, this limitation will have less impact on recall but it will never be ixed
entirely.
Botles, bags, cans, and jars have a large showing in the false negative category due to
Hough line detection difficulties on curved lines (see Figure 15 ) . One possibility to get around
this limitation is a more rigorous line detection step in which a segmented HT is performed
and regions which contain connecting detected lines are grouped together. These grouped re-
gions could be used to warp a curved image into a rectangular area for further analysis.

FIGURE 15 NL with curved lines.

Smaller grocery packages (see Figure 16 ) tend to have irregular NLs that place a large
amount of information into tiny spaces. NLs with irregular layouts present an extremely dif-
icult problem for analysis. Our algorithm beter handles more traditional NL layouts with