Method of reconstructing a high resolution image by tomodensitometry

Claim 1

A method of reconstructing images by tomodensitometry, of the type having a source-detectors assembly, the method comprising rotating a source-detectors assembly in the plane of a section whose image it is desired to form, said source emitting a penetrating radiation beam in the form of a fan and taking and storing a succession of views for different rotational angles β of the source-detectors assembly with respect to a fixed reference axis system (XOY) of said plane, a plurality of detectors being arranged so as to converge towards the focal point (F) of said source and each view being sampled for angles γ representative of the detectors and referenced from a straight line (OF) joining a center of rotation (0) of the source-detectors assembly and said focal point (F), an angular shift between two adjacent detectors Δγ being constant and an angular shift between views being constant, with the angular shift between said straight line (OF) and the nearest detector thereto adjusted to ##EQU8## and taking and storing a series of views spaced apart over a complete revolution of said source-detectors assembly, measuring for each view a number of intermediate values representing sampling interlaced with that of said detectors, each intermediate value of a given angle γ being obtained by a combination of at least two linear absorption values inferred from measured values taken from the views situated on each side of a fictitious view whose image by a transformation T (θ,γ)=(θ+(2k-1)π,-γ) contains a point representative of said intermediate value in a space [θ,γ] with θ=β+γ and k a relative integer, said measured values of said views being taken from an angle detector opposite one of said desired intermediate value, and applying a reconstruction algorithm, comprising a convolution and a rear projection to the views thus completed by said intermediate values.

Claim 2

A method of reconstructing images by tomodensitometry, of the type having a source-detectors assembly, the method comprising rotating a source-detectors assembly in the plane of a section whose image it is desired to form, said source emitting a penetrating radiation beam in the form of a fan and taking and storing a succession of views for different rotational angles β of the source-detectors assembly with respect to a fixed reference axis system (XOY) of said plane, a plurality of detectors being arranged so as to converge towards the focal point (F) of said source and each view being sampled for angles γ representative of the detectors and referenced from a straight line (OF) joining a center of rotation (0) of the source-detectors assembly and said focal point (F), an angular shift between two adjacent detectors Δγ being constant and an angular shift between views being constant, with the angular shift between said straight line (OF) and the nearest detector thereto adjusted to ##EQU8## and taking and storing a series of views spaced apart over a complete revolution of said source-detectors assembly, measuring for each view a number of intermediate values representing sampling interlaced with that of said detectors, each intermediate value of a given angle γ being obtained by a combination of at least two linear absorption values inferred from measured values taken from the views situated on each side of a fictitious view whose image by a transformation T (θ,γ)=(θ+(2k-1)π,-γ) contains a point representative of said intermediate value in a space [θ,γ] with θ=β+γ and k a relative integer, said measured values of said views being taken from an angle detector opposite one of said desired intermediate value, and applying a reconstruction algorithm, comprising a convolution and a rear projection to the views thus completed by said intermediate values.