| Summary: | 碩士 === 長庚大學 === 商管專業學院碩士學位學程在職專班資訊管理組 === 105 === With the evolution of information technology, traditional signage has been gradually replaced by digital signage. Advertisers invest a lot of manpower and cost to evaluate the advertisement effectiveness. The audience rating is the key index. Although face recognition systems are emerging, but the cost is still very high. Manual inspection remains the common approach to audience rating.
This study attempts to utilize lower cost technology to assist ratings surveys for digital signage. A smart phone with built-in three-axis acceleration sensor and Bluetooth and database was used plus iBeacons. We developed a mobile App to collect iBeacon UUIDs (to understand whether users enter digital signage areas), the distances from the iBeacons (whether users are close to digital signage) and the walking time of each step (how long users stay in front of digital signage).
In order to allow the pedometer to be placed in the pocket or in the hand, the acceleration vector XYZ was used to calculate the vector intensity (SVM). The filter of Butterworth low-pass filter was used. The peak detection was used to determine steps. We used threshold values for jitter removal.
In order to check the correctness of the designed App, we conducted seven experiments to test the accuracy of the pedometer. The average accuracy rate of the seven experiments was 98.79%. Moreover, we also conducted experiments to determine the accuracy of detecting iBeacon areas. The findings revealed iBeacon areas were all correctly detected.
|
|---|
