Researchers pitch combined biometrics to enable identification from blurred, dark images

As demand for remote identification via surveillance increases, a survey of academic work on biometrics including facial recognition and gait recognition finds the gaps in understanding for dealing with images that are blurred, poorly-lit, from a difficult angle or where the subject has been partly cut from the frame or is simply too small. Ways to improve methods and lighten algorithms that will help bring better recognition to the edge are called for.

‘A Survey on Face and Body Based Human Recognition Robust to Image Blurring and Low Illumination’ from the Division of Electronics and Electrical Engineering, Dongguk University in Seoul, published by MDPI, centers on overall biometric ‘human recognition’ from problematic images and attempts at improving image quality to counteract the issues.

While facial recognition has received much of the research attention, as faces are deemed to contain the most important information for identification, the researchers argue that multimodal body and gait recognition can help with what they term overall ‘human recognition.’

Low-resolution images have been sufficiently dealt with, yet survey papers on blurred images are not comprehensive, which the paper attempts to address. It reviews studies on blurred image restoration and low-illumination and classifies them as to whether or not deep learning was used and whether face and body were combined.

The team tackles indoor and outdoor settings which generate distinct problems. Indoor images are more prone to motion blur and difficult angles as the subjects are closer. Outdoors, illumination can be non-uniform and images lower in resolution.

“No study has yet been conducted on body-based human recognition robust to image blurring in indoor environments; in this case, only the body region is used, dismissing the face. In other words, neither body-based recognition nor body-based re-identification have been studied yet,” states the study. “This is because, compared to the face region, the body region requires more global features for recognition. This in turn implies that the recognition performance is not significantly affected by image blurring.”

The survey covers how the degree of blurring is evaluated through image quality assessment, how the color of clothing can affect results for body-based recognition. Gait recognition can be used in more situations due to the low impact of image blurring.

Further studies are needed on the impact of low-illumination on gait-based recognition. Further studies are also required for human recognition in more severe instances of low lighting. Studies so far have shied away from this as “it is difficult to restore colors perfectly when converting severe low-illumination images into normal-illumination images. It is expected that these problems could be solved through the various deep learning methods.”

“As a whole, both face- and body-based recognition shows a higher accuracy and more processing time than the face-based method,” state the authors, who hope to help biometrics researchers as more demands are made for criminal and missing persons detection, as well as settings such as driver identification in vehicles and crowd analytics.

Article Topics

biometric identification  |  biometrics  |  biometrics research  |  facial recognition  |  gait recognition  |  multimodal biometrics  |  re-identification  |  video surveillance