564 BELL SYSTEM TECHNICAL JOURNAL 



liition of the image can readily be varied over a substantial range, and 

 with provision for making the horizontal resolution different from the 

 vertical. The use of motion pictures instead of actual television 

 images permits sharpness to be studied independently of other factors, 

 and facilitates the experimental procedure. 



The relationship between the television image and the motion 

 picture which simulates it will be determined on the basis of their 

 subjective equality in sharpness. For that purpose, a television image 

 reproduced by an apparatus * of known characteristics is to be com- 

 pared with a projected out-of-focus motion picture of the same scene, 

 under the same conditions of size, viewing distance, brightness and 

 color. (The motion picture will in general be superior in the rendition 

 of tone values and in respect to flicker, and will of course not show the 

 scanning line structure of the television image or any of the degrada- 

 tions commonly encountered in electrical transmission.) When the 

 two images are judged to be equally sharp by the median one of a group 

 of observers, the size of the figure of confusion of the motion picture is 

 to be taken as the measure of the resolution of the compared television 

 image. 



The figure of confusion of the motion picture is that small area of 

 the projected image over which the light from any point in the film is 

 spread. Every point produces its own figure of confusion, of pro- 

 portionate brightness, and the overlapping of the figures in every 

 direction accounts for the loss of sharpness. When the projection lens 

 is "in focus," the figure of confusion is a minimum one set by the aberra- 

 tions of the optical system and by diffraction effects. As the lens is 

 moved away from the "in focus" position, the figure of confusion be- 

 comes larger and assumes the shape of the aperture stop of the projection 

 lens. If the illumination of the aperture stop is uniform, this larger fig- 

 ure of confusion is a well-defined area of uniform brightness. We used 

 a rectangular aperture stop, at the projection lens, whose height and 

 width could be varied reciprocally so as to maintain constant area of 

 opening, and we used a calibrated microscope to measure the departure 

 of the lens from the "in focus" position. Thus we could produce images 

 of various degrees of sharpness and of unequal horizontal and vertical 

 resolutions. 



This method of specifying the resolution of an image in terms of the 

 size of the figure of confusion affords an important advantage. It 

 avoids the necessity for postulating any particular relation between 

 the resolution and the spatial distribution of brightness values about 



* The television apparatus comprised a mechanical film scanner and an electronic 

 reproducing tube designed specifically for television. A description of it is given in 

 reference 9. 



