Methods of evaluating images of gamma-ray radiography were investigated from the standpoint of the optical transfer function in receiving system with different sources. The optical transfer function was expressed by the sum of two error functions having different dispersion factors. The evaluation methods from several categories in terms of optical transfer function; that is, the function's value at a fixed spatial frequency, the spatial frequency at a given attenuation value, area, moment and others are used respectively for the evaluation. Also the relationships between various sharpness indexes usually used and the optical transfer functions were studied. The resultant value obtained by each evaluation method varied with each source. The ratios of the values obtained with three sources, Ir-192, Cs-137 and Co-60, were 135 : 100 : 73. If an object is inserted in the system, these ratios will be changed. Problems concerning the grains of metallic silver from film emulsion and the optical transfer function of observation itself are not considered in this paper.
The experimental results of the activation with polychromatic fast neutrons produced by9Be (d, n) 10B reaction in a 2 MV Van de Graaff accelerator are presented. These neutrons, although they do not have much energy to produce the usual fast neutron reactions, do induce usable (n, n') and (n, γ) reactions. Sensitivity values of various elements, and analytical applications of the Be-D neutron activation are presented.
For on-off type and follow-up type level gauges, the primary and scattered beams of gamma-rays were limited only to a horizontal plane between lead blocks in order to increase the counting rate from a given source without deterioration of the resolving power of level. The gamma-ray source used was137Cs, and the detector was a NaI (Tl) scintillator 44mm in diameter and 25mm thick shielded not only from the direct beam but also from the photons scattered once by the vessel walls. They were arranged on a horizontal plane. The vertical gaps between the lead blocks used were 10, 20 and 30mm. It has been found that the width of the gap affects both the counting rate for a full vessel and the relative sensitivity at the level to be detected. The former is affected more by the gap width on the detector side than that on the source side, and the latter vice versa. It is proposed that merits of these kinds of level gauges may be expressed by the minimum product of the radioactivity of the source and the time constant of the rate meter circuit, necessary to respond to a level change of unit length. From the values, it has been found that the above-mentioned arrangements are better than those using a narrow beam reported in a previous paper.