Apparatus

Data Collection

In an effort to improve on previous work done on this subject, it was agreed upon to construct an apparatus from scratch. This new apparatus was constructed with several design goals in mind. First, it was to stand upright and be placed on a lab counter for ease of video recording. Next, adjustability was kept in mind through the use of an actual guitar tuner for frequency selection, a sliding left-hand wooden panel to adjust string length, and easy string installation in order to test various string materials. Figures 2 and 3 show this apparatus.

Several background boards were used for video calibration in pursuit of proper contrast to the suspended string. Although a black background is pictured, it was found that the white background board was better suited for the contrast in mind. A Phillips T12 40-Watt light strip (not pictured) was suspended by the top metal bar of the apparatus and aimed downwards onto the backdrop. This gave a bright background in the recording and a silhouette of the string when taking video from the front of the apparatus. This produced a good contrast and simplified image processing.

Figure 2

Figure 3

Image Processing

Several computer applications were used to complete video and image processing for this project. Raw video files were taken using the Casio EX-F1 DSLR camera at 1,200 frames per second. These video files were imported into Adobe Premiere Pro for video processing. The videos were cropped to proper length,as the high-speed video exaggerated any gap between recording initialization and string plucking. Each video was also changed to gray scale coloring in order to simplify further analysis. Still frames from each video were then exported from the application for one full period of wave motion. This translated to approximately 20 image frames per period.

These exported still frames were then imported into ImageJ image processing software. ImageJ was used in order to get a simplified, high contrast image of the string. In this software, the image was converted to an 8-bit image file. Then the “Subtract Background” function was used to achieve a more uniform light background. Brightness and contrast settings were then modified to darken the string pixels on the newly modified background. After these steps, a high contrast image of the string began to take form. Finally, the “Threshold” values of the image were modified. This further pixelated the image and gave it the highest contrast seemingly possible.

All of this processing led to using ImageJ’s “Save X and Y Coordinates” function. With the newly processed images, it was simple for the program to detect the background color value and save X-Y coordinates of all pixels that vary from it. This gave detailed text files consisting of X and Y coordinates as well as associated color values for each pixel that differed from the background. Using these text files, it was possible to construct a plot of the string’s position in space at a specific moment in time. This process is summarized in Figure 4 below.

Figure 4