Experimental X-Ray Ghost Imaging

Abstract

We report an experimental proof of principle for ghost imaging in the hard x-ray energy range. We used a synchrotron x-ray beam that was split using a thin crystal in Laue diffraction geometry. With an ultra-fast imaging camera, we were able to image x-rays generated by isolated electron bunches. At this time scale, the shot noise of the synchrotron emission process is measurable as speckles, leading to speckle correlation between the two beams. The integrated transmitted intensity from a sample located in the first beam was correlated with the spatially resolved intensity measured in the second, empty, beam to retrieve the shadow of the sample. The demonstration of ghost imaging with hard x-rays may open the way to protocols to reduce radiation damage in medical imaging and in non-destructive structural characterization using Free Electron Lasers.Ghost imaging, in its basic form, is the technique of indirectly imaging a sample by using the correlation between the intensity recorded at two detectors illuminated by spatially separated correlated beams [1]. A bucket detector measures the total intensity transmitted (or scattered) by a sample, placed in one of the beams. The sample image is then retrieved by correlating the output of the bucket detector with a pixel array detector located in the other beam, namely the one that has not directly interacted with the sample. Initially demonstrated with entangled photon pairs [2], ghost imaging was subsequently performed using correlation between classical coherent light beams [3]. The protocol was shown to be very robust, leading to experimental studies on ghost imaging using pseudo-thermal light [4][5][6], true thermal sources [7], and eventually computational ghost imaging [8], where a computer-controlled spatial light modulator generates a series of known illuminating fields, altogether removing the need for imaging the empty beam. Of relevance for this paper is also a very recent demonstration of Fourier transform ghost imaging using speckle fields generated with partially coherent synchrotron x-rays [9]. At the heart of thermal ghost imaging is the speckle correlation in the intensity fluctuations of the illuminating beam. The speckles can be produced either by nearfield diffraction of a coherent beam by a slowly moving diffracting object [4-6, 9], or taking advantage of the natural fluctuations of true thermal light [7], as in the Hanbury Brown-Twiss (intensity) interferometer [10]. In this Letter we use the latter mechanism to produce the first proof of principle demonstration of hard x-ray direct ghost imaging using synchrotron emission from an undulator in a third generation synchrotron storage ring. Synchrotron emission from an ultra-relativistic electron bunch provides a natural thermal source of hard x-rays. Intensity correlation x-ray experiments, proposed as far back as 1975 [11] (see also [12]), were employed several times for coherence characterization of synchrotron [13][14][15] and x-ray Free Electron Laser (FEL) [16] beams. To date though, x-ray spec...

Statistical properties of a free-electron laser revealed by Hanbury Brown–Twiss interferometry

“…This could lead to quantum optics experiments such as the exploration of nonclassical states of light [57], superresolution and quantum imaging experiments [58,59], or ghost imaging experiments [60,61] at the FEL sources. As it was shown in our work, by sufficient monochromatization, a single longitudinal mode of FEL radiation or Fourier limited pulses with high photon flux can be achieved.…”

On the “unreasonable” effectiveness of transport of intensity imaging and optical deconvolution

“…(1)- (3) is defined in terms of the correlation of light irradiance at different spatial positions, a situation which has obvious parallels, for example, with the Hanbury Brown -Twiss experiments [13][14][15][16][17][18] and with recent studies in X-ray ghost imaging [19][20][21][22]. Note also that, due to our choice of variables in eq.…”

Perspectives for Applications of Quantum Imaging

“…This is relevant for objects that are sensitive to light, for example, in the biological and life sciences. It can be especially useful when measuring at short wavelengths, especially in the UV spectral range and below, where classical GI was already demonstrated for X‐rays . The number of needed photons may be further reduced by using techniques as compressed sensing, which faithfully reconstruct images from an incomplete amount of measured information.…”