Modulating mobile-device displays based on ambient signals to reduce the likelihood of fraud

1. A tangible, non-transitory machine-readable media storing instructions to classify ambient signals to reduce fraudulent use of stored value card information, wherein the instructions, when executed by one or more processors of a mobile computing device effectuate operations comprising:

receiving, with the mobile computing device, from a remote server, balance-access information by which a stored value card balance can be spent at a point of sale terminal;

storing, with the mobile computing device, the balance-access information;

determining a geolocation of the mobile computing device;

in response to determining that the geolocation of the mobile computing device has crossed a geo-fence associated with a retail establishment, requesting store-specific parameters from a remote server;

receiving the store-specific parameters;

sensing, with one or more sensors of the mobile computing device, ambient signals;

classifying the ambient signals, based on the store-specific parameters, as indicating the user is in the retail establishment; and

in response to the classification, displaying, on a display screen of the mobile computing device, the balance-access information, such that the balance-access information can be input to a point-of-sale terminal.

2. The media of claim 1 , wherein:

sensing ambient signals comprises:

sensing audio;

sensing an image;

sensing an orientation; and

sensing a wireless beacon; and

classifying the ambient signals as indicating the user is in the retail establishment comprises:

calculating a score based on a weighted combination of a classification of the sensed audio, a classification of the image, a classification of the orientation, and a classification of the wireless beacon; and

determining that the score satisfies a threshold.

3. The media of claim 1 , wherein determining a geolocation of the mobile computing device comprises:

receiving an encrypted rolling code emitted by a wireless beacon; and

validating that the encrypted rolling code corresponds to wireless beacon of a retailer.

4. The media of claim 1 , wherein sensing ambient signals comprises sensing audio with a microphone of the mobile computing device.

5. The media of claim 4 , wherein classifying the ambient signals comprises:

normalizing the audio;

extracting a feature vector from the normalized audio; and

scoring the feature vector with a value indicating a likelihood of whether the user is in a retail establishment.

6. The media of claim 4 , wherein classifying the ambient signals comprises:

extracting a feature by passing a representation of the audio through a band-pass filter.

7. The media of claim 4 , wherein classifying the ambient signals comprises:

extracting a feature by determining a Fourier transform of a representation of the audio.

8. The media of claim 4 , wherein the features comprise:

an audio signal within a range of frequencies; and

duration of the audio signals within the range of frequencies.

9. The media of claim 8 , wherein the range of frequencies is selected from among a plurality of ranges of frequencies based on a geolocation sensed by the mobile computing device.

10. The media of claim 1 , wherein sensing ambient signals comprises receiving a reading from an inertial measurement unit (IMU) of the mobile computing device.

11. The media of claim 10 , wherein classifying the ambient signals comprises:

determining an orientation of the mobile computing device with respect to gravity.

12. The media of claim 10 , wherein classifying the ambient signals comprises:

classifying a multi-dimensional time series of readings from the IMU as indicating a gesture.

13. The media of claim 12 , wherein classifying a multi-dimensional time series of readings from the IMU as indicating a gesture comprises dynamic time warping the multi-dimensional time series.

14. The media of claim 1 , wherein sensing, with one or more sensors of a mobile computing device, ambient signals comprises:

sensing an image with a camera of the mobile computing device.

15. The media of claim 14 , wherein classifying the ambient signals comprises:

classifying the image as containing at least part of a point-of-sale terminal.

16. The media of claim 14 , wherein classifying the ambient signals comprises:

classifying the image as containing at least part of a point-of-sale terminal by detecting translation invariant features in the image corresponding to the point-of-sale terminal with a convolution layer of a neural network.

17. The media of claim 14 , wherein classifying the ambient signals comprises:

classifying two images from two cameras of the mobile computing device by determining that one image from a camera facing in a direction opposite the display screen contains the user's face, such that the display screen is oriented away from the user.

18. The media of claim 1 , wherein classifying the ambient signals comprises performing steps for classifying the ambient signals.

19. The media of claim 1 , wherein displaying, on a display screen of the mobile computing device, the balance-access information comprises:

receiving an input from a touchscreen of the mobile computing device indicating a user input;

in response to the user input, displaying the balance-access information; and

determining that the user input has ceased and, in response, ceasing to display the balance access information.