In this paper we present a system for preserving the privacy of individuals in a video surveillance scenario. While a person's privacy should not be revealed to a viewer of the video without special needs, it is still important that the action in a scene as the semantic content of a video remain perceivable by a human observer. The proposed system uses edge detection and adaptive thresholding in order to estimate the persons' silhouettes in a video scene and thus rendering most of their actions visible, while hiding sensitive personal information. In order to obtain a more complete contour around a person, an adaptive thresholding scheme using edge histograms is used as well as background subtraction which limits the edge extraction to foreground masks and thus avoids distraction of the viewer's eyes to background structures.
With the increasing usage of Video Surveillance systems
and a continuously growing amount of CCTV data being
recorded and looked at, the need to apply privacy protection
techniques in this area rises as well. In order to increase
acceptance of CCTV-cameras among people being observed,
it is of special importance to ensure their personal rights be
not violated, while the surveillance system is still able to
work and security staff is able to identify critical events in
a video stream. The MediaEval 2013 Visual Privacy Task
[
For obfuscation of personal data, we propose to show only the contour of a person in the video. In case of a low number of persons in the scene, their movements and actions could be identified while it is still impossible to identify personal details such as the person’s face, color of clothes or skin color. In this paper, we show a system which uses adaptive edge detection to determine the contour of a person and in the privacy-filtered result video replaces the interior of a person’s silhouette by background information. 2.
The proposed system uses a background model in order
to exchange the information within a person’s contour by
recent background information (see Fig. 1). The
bounding boxes which are provided as manually-annotated ground
truth for the purpose of MediaEval are used in order to
identify where the privacy filter needs to be applied in the image.
In these regions, edge detection is performed, and the result
is blended into the background image. As the bounding
boxes for objects could also be provided automatically (e.g.
by human detection algorithms or trackers such as [
Our algorithm uses two background models: a) a standard
Gaussian-Mixture Model (GMM) similar to [
BGi(x, y) = (1 − α) · BGi−1(x, y) + α · I(x, y) (1)
In order to adapt the background model quickly to the new frame, we use rather high values for α (e.g. α = 0.35) and update this model every 5th frame. This also has the advantage that the system can cope quickly with slight camera movements. In some of the PEViD videos, the camera shakes a bit and the image is shifted for some pixels. While first the background image in our system gets a bit blurry in these situations, after a few frames our system maintains the current background and is not disturbed by older values anymore. The GMM uses a much slower learning rate and is only used to mask out background parts within the given region of interest (ROI) - a task for which the previously described background model b) alone would be to simple. 2.2
We use Canny edge detection [
In our system which has to work under varying conditions (indoor, outdoor, different lighting, changing weather...), we propose to adapt the threshold according to the gradient histogram in the ROI. For the given ROI, we compute the absolute values of the gradient information and set up a histogram over every ROI. Using the assumption, that most of a human’s silhouette can be recovered by the highest 15 % of the gradient information, we can thus find T1 for the Canny algorithm. In order to close holes in the silhouette, T2 is set to T2 = 0.9 · T1. This automatic choice of T1 and T2 adapts well to most scenes in the data set and gives a good contour information of a walking person in most cases.
The evaluation of the proposed filter is based on
objective and subjective metrics described in [
In regards to the intelligibility and appropriateness the results of the subjective and objective evaluations related to the average of all participants are contrary. While for these categories the score of the objective results are less than the average, the score of the subjective results is higher than the average score. As the proposed filter extracts and reprojects the edge information into the image and fills the interior of
0.693835 0.368627
Adaptive edge filter 0.678333 0.683750 0.532500 a person with background information, it is obvious that a standard person detection and tracking method which is usually based on extracting edges by HoG or wavelet filtering and is trained on normal videos cannot have the same performance as in unfiltered images. To apply the filtered videos for automated video analytics the respective preprocessing of the methods has to be adapted and we thus recommend a new training step based on feature extraction on the filtered images. However, based on the results of the subjective evaluation, the filter is especially appropriate for display in video surveillance systems that are guided and evaluated by human operators. 4.
In this paper we showed how adaptive edge detection can be used to preserve the privacy of people in CCTV videos. The subjective test demonstrates that the proposed filter outperforms most of the other privacy filters proposed in the MediaEval 2013 challenge. Thus, our method is of special interest for semi-automatic video surveillance systems which need to hide sensitive personal information while preserving the context of actions.
This work was supported by the European Commission under contracts FP7-261743 VideoSense.