---
title: 'HiXray: Security Benchmark & Diffuse Scanning'
url: https://www.emergentmind.com/topics/hixray
type: topic
---

# HiXray: Security Benchmark & Diffuse Scanning

HiXray is a polysemous term in recent X-ray imaging literature. In security inspection, it denotes a high-quality benchmark of real-world airport baggage X-ray images for prohibited-item detection, introduced together with the Lateral Inhibition Module (LIM) for boundary-aware detection under overlap and clutter. In computational X-ray imaging, it also denotes a high-resolution scanning method that uses a diffuse, Huffman-patterned probe to spread dose spatially and recover sharp images by deconvolution. These two usages are technically unrelated, but both have become reference points in their respective subfields [2108.09917] [2410.18348].

## 1. Terminological scope and research contexts

In the security-imaging literature, HiXray was introduced as a **High-quality X-ray security inspection image dataset** containing real-world airport scans and professional bounding-box annotations for prohibited items [2108.09917]. Subsequent comparative studies use HiXray as one of the standard public benchmarks for X-ray illicit-object detection, typically alongside datasets such as EDS and PIDray [2505.00564].

In a distinct line of work, HiXray is the abbreviated name of **“High-resolution x-ray scanning with a diffuse, Huffman-patterned probe to minimise radiation damage”**, a coded-scanning framework in which a broad structured beam is rastered across an object and the resulting bucket signals are decoded to reconstruct a sharp image [2410.18348].

A common source of confusion is to treat all occurrences of “HiXray” as referring to the same system. The supplied literature does not support that usage: one body of work concerns airport security inspection and object detection, while the other concerns coded-aperture-like X-ray scanning and dose management.

## 2. HiXray as a real-world security-inspection benchmark

The HiXray dataset was collected from the carry-on luggage of an international airport’s security X-ray machines, and every image is a genuine scan rather than a staged or synthetic overlay [2108.09917]. It contains **45,364 RGB X-ray scans** and **102,928 bounding boxes** across **8 categories**, with **2.27 objects per image** on average. The classes are **PO1** (portable charger 1, a prismatic lithium-ion cell), **PO2** (portable charger 2, a cylindrical lithium-ion cell), **WA** (water), **LA** (laptop), **MP** (mobile phone), **TA** (tablet), **CO** (cosmetic), and **NL** (nonmetallic lighter) [2108.09917] [2507.17508].

The annotation format is axis-aligned bounding boxes with class labels. The annotations were drawn by professional airport security inspectors following a Pascal-VOC-style protocol with exhaustive labeling, consistency checks, and cross-inspection [2108.09917]. The image data are described as pseudo-color RGB images from a dual-energy X-ray scanner, single-view, high-quality captures collected at an international airport [2507.17508]. Reported image sizes are JPEG with average **1,200 × 900 px**, up to **2,000 × 1,040 px** [2108.09917].

The benchmark’s visual domain is materially encoded. The dataset description explicitly notes color-by-material mapping with **orange for organics, blue for metals, and green for mixtures** [2108.09917]. In later evaluations, the dataset is characterized as **single-scanner data** with **no explicit domain shift within HiXray**, and with **moderate clutter in real baggage scans** but **no deliberately hidden items**, unlike PIDray [2505.00564]. This matters because HiXray is used less as a benchmark for adversarial concealment than as a benchmark for fine-grained detection under realistic overlap, everyday baggage clutter, and material ambiguity.

## 3. LIM: boundary-aware feature refinement on HiXray

The paper that introduced HiXray also introduced the **Lateral Inhibition Module (LIM)**, a plug-in feature module intended to improve prohibited-item detection in X-ray images by suppressing irrelevant information and emphasizing identifiable boundaries [2108.09917]. LIM is described as comprising a **Bidirectional Propagation (BP)** submodule and a **Boundary Activation (BA)** submodule. It can be inserted between a convolutional backbone and the detection head of one-stage or anchor-free detectors such as SSD, YOLOv5, and FCOS [2108.09917].

For the top-down BP pathway, if \(L\) is the total number of feature levels and \(\mathcal{F}^l(x)\) is the feature map at level \(l\), the reported formulation is

\[
A^{l} \;=\;\mathcal{V}\bigl(\mathcal{F}^{l}(x)\bigr)
\;+\;\sum_{m=1}^{L-l}\mathcal{U}^{m}\bigl(A^{l+m}\bigr)\,.
\]

After BA, the bottom-up BP pathway is written as

\[
C_t^{\,l}
\;=\;
\mathcal{V}\bigl(B^{l}\bigr)
\;+\;\sum_{m=1}^{l-1}\mathcal{D}^m\bigl(C_t^{\,l-m}\bigr),
\qquad
C^{l}
\;=\;
C_t^{\,l} \;+\;\mathcal{F}^{l}(x)\,.
\]

The BA submodule applies directional running-max operations to emphasize boundary structure. For the left-to-right pass on \(A^l\in\mathbb{R}^{H\times W\times C}\), the paper gives

\[
B^{l}_{ijc}
=
\begin{cases}
A^{l}_{iWc}, & j = W, \\[6pt]
\max\{A^{l}_{ijc},\, A^{l}_{i,j+1,c},\,\dots,\,A^{l}_{iWc}\}, & \text{otherwise.}
\end{cases}
\]

The interaction of BP and BA is described in explicitly neuro-inspired terms: the top-down path filters spurious activations, BA re-activates salient edge cues, and the bottom-up path propagates the refined features while suppressing context noise [2108.09917].

Quantitatively, the reported gains on HiXray are measured by **mAP@0.5 IoU**. On HiXray, **SSD** improves from **71.4%** to **73.1%**, **FCOS** from **75.3%** to **76.9%**, and **YOLOv5s** from **78.6%** to **80.1%** when LIM is added. The paper further states that, against the prior specialized module **DOAM** on the same backbones, LIM consistently outperforms by approximately **0.6–2.6 percentage points** [2108.09917].

## 4. Comparative detector behavior on HiXray

Later studies situate HiXray within a broader benchmarking framework for X-ray illicit-object detection. The common detection metrics are based on Intersection over Union,

\[
\mathrm{IoU}(A,B)\;=\;\frac{\lvert A\cap B\rvert}{\lvert A\cup B\rvert}\,,
\]

and on mean average precision variants such as **mAP@50** and **mAP@[50:95]** [2507.17508]. In a focused hybrid-architecture study, the HiXray protocol uses an **80% training / 20% testing** split, **batch size 18**, and **SGD** with **learning rate \(0.01\)** and **momentum \(0.937\)**; early stopping is based on validation loss, and data augmentation follows the default Ultralytics routines of random flips, mosaic, and mixup [2505.00564].

A later large-scale comparison reports the following HiXray results for ten detectors [2507.17508]:

| Detector | mAP@50 | mAP@[50:95] |
|---|---:|---:|
| YOLOv8 + CSPDarkNet53 | 0.845 | 0.564 |
| YOLOv8 + HGNetV2 | 0.833 | 0.557 |
| YOLOv8 + CHR | 0.811 | 0.523 |
| YOLOv8 + DOAM | 0.830 | 0.545 |
| YOLOv8 + LIM | 0.828 | 0.525 |
| DINO + Swin-B | 0.849 | 0.535 |
| Co-DETR + Swin-B | 0.857 | 0.531 |
| RT-DETR + HGNetV2 | 0.839 | 0.510 |
| YOLOv8 + Next-ViT-S | 0.841 | 0.551 |
| RT-DETR + Next-ViT-S | 0.818 | 0.483 |

These numbers establish a nuanced ranking. The **highest mAP@50** is achieved by **Co-DETR** at **0.857**, whereas the **highest mAP@[50:95]** is achieved by **YOLOv8 + CSPDarkNet53** at **0.564** [2507.17508]. In the more targeted hybrid-architecture evaluation, the same YOLOv8 baseline with its default backbone is reported to outperform all hybrid variants on HiXray, including object-level **mAP\(^{50:95}\)** for every class [2505.00564].

The stricter localization metric therefore favors the CNN-only baseline on HiXray, even though a transformer detector attains the best single-threshold **mAP@50**. This suggests that, on this dataset, local appearance modeling and localization fidelity remain particularly consequential.

## 5. Dataset-specific challenges and empirical interpretation

The comparative papers describe HiXray as a **homogeneous single-scanner dataset** with **no large cross-domain shifts**, in contrast to EDS [2505.00564] [2507.17508]. This has direct architectural consequences. One study concludes that on **HiXray-like scenarios** the recommended choice is **YOLOv8 with CSP-DarkNet53 backbone** for the highest mAP and inference speed, and that hybrid CNN-transformer backbones should be reserved for cases with known domain shifts [2505.00564].

Class-level behavior is also uneven. The later comparative evaluation reports that the **best detected classes** are **high-density, geometrically simple items**, such as **laptops** and **portable chargers**, while the **worst detected classes** are **low-density or highly variable items**, such as **water**, **cosmetic**, and **non-metallic lighter** [2507.17508]. It further states that **portable charger 1** and **portable charger 2** are often confused because of similar shape and material, and that **fine-grained electronics** such as phones and tablets benefit from local feature modeling in CNNs more than from global-context transformers [2507.17508].

Object size is not the dominant factor in relative model ranking, but it is still a difficulty axis. The object-size analysis on HiXray uses COCO definitions of small, medium, and large objects and reports that **small objects** show a **slight dip of approximately 5–7 points** relative to medium and large objects, whereas medium and large objects show nearly identical performance across detectors; the variance across scales is reported as **less than 3%** between the best and worst models [2505.00564]. The same study concludes that small objects remain the most challenging.

Another important domain property is that HiXray does **not** define explicit occlusion subsets, yet ordinary baggage overlap remains common: chargers may be on phones, cables may cross devices, and low-\(Z\) materials such as water and plastics produce faint signals and higher false-negative rates [2507.17508]. The difficulty is therefore not simply heavy clutter, but the conjunction of overlap, material attenuation variability, and fine-grained inter-class similarity in a relatively clean single-scanner domain.

## 6. HiXray as diffuse Huffman-patterned X-ray scanning

In a separate research program, HiXray denotes a high-resolution X-ray scanning strategy that replaces a tightly focused probe with a broad, structured probe derived from **Huffman sequences** [2410.18348]. The starting point is a 1D Huffman sequence \(H_L\) of length \(L\), whose aperiodic autocorrelation

\[
R_H(\tau)\;=\;\sum_{i=0}^{L-1-\tau}H_L(i)\,H_L(i+\tau)
\]

is as close as possible to zero for all nonzero shifts except for two inevitable end terms. The guiding idea is that

\[
H_L\otimes H_L\;\approx\;\delta,
\]

with \(\otimes\) denoting aperiodic correlation [2410.18348].

A 2D mask is formed by an outer product, \(H_{2D}(i,j)=H_L(i)\,H_L(j)\), yielding a 2D pattern with strongly \(\delta\)-like aperiodic autocorrelation [2410.18348]. Because raw integer Huffman sequences can have a very large dynamic range, the authors report the use of controlled scaling, integer rounding, and small random dither perturbations to compress the entries to \(\{-3,\dots,+3\}\). Quality is monitored using **RMS**, **mean absolute value**, **zero-fraction \(f^z\)**, **merit factor \(M^f\)**, **peak-to-sidelobe ratio \(R^0\)**, **spectral flatness \(d^F\)**, and the **condition number \(\kappa\)** of the convolution operator [2410.18348].

The masks are fabricated on **SiO\(_2\)** wafers with sputtered **tantalum (Ta)** up to approximately **5 \(\mu\)m** thick. Two implementations are reported. **Binary masks** use a single 5 \(\mu\)m Ta layer and encode gray values by subdividing each pixel into \(3\times3\) subpixels, with pixel pitches of **8 \(\mu\)m**, **10 \(\mu\)m**, or **15 \(\mu\)m**. **Quaternary masks** realize four transmission levels \(\{0,1,2,3\}\) through three successive lithography-plus-etch steps, with pixel pitches of **10 \(\mu\)m**, **15 \(\mu\)m**, or **20 \(\mu\)m** [2410.18348].

Validation proceeds in two ways. First, the transmitted beam is imaged directly using a uniform monochromatic synchrotron beam: **12.4 keV** for quaternary masks and **20 keV** for binary masks, with a **6.5 \(\mu\)m** pixel detector [2410.18348]. Second, a small pinhole of **5–20 \(\mu\)m** is rastered across the mask, and bucket signals are formed from the full-field detector image. For a **32×32 quaternary mask** and a **20 \(\mu\)m pinhole**, the resulting **32×32 bucket map** recovers all seven gray levels, and its cross-correlation with the ideal mask has a sharp central peak with **100%** at the center and near zero elsewhere [2410.18348].

The forward model for scanning an object \(O(x,y)\) with a known pattern \(H(i,j)\) is

\[
b(x,y)\;=\;\sum_{i,j}H(i,j)\,O(x+i,y+j)\;+\;n(x,y)
\;\equiv\;(H\ast O)(x,y)+n,
\]

or, in matrix form, \(b=A\,x+n\) [2410.18348]. Deblurring can then be performed by cross-correlation with the mask:

\[
\widehat O\approx b\otimes H
\quad\Longrightarrow\quad
\widehat O\approx O\ast(H\otimes H)\approx O.
\]

The same paper notes that one may also write \(x=A^{-1}b\) or use Tikhonov-type regularization \((A^TA+\lambda I)^{-1}A^T b\) [2410.18348].

## 7. Performance, dose redistribution, and limitations of the diffuse-probe method

The spatial resolution of the diffuse-probe HiXray method is set by the **mask pixel pitch** and the **scan step size**, both reported in the range **8–20 \(\mu\)m** in the experiments [2410.18348]. A **32×32** mask with **20 \(\mu\)m** pitch therefore yields a **32-pixel field of view** with **20 \(\mu\)m** resolution over a **640 \(\mu\)m** span [2410.18348]. The numerical conditioning is reported as **\(\kappa\lesssim1.5\)** for 2D masks up to **32×32**, which the authors present as ensuring stability against noise and misalignment [2410.18348].

The method’s central motivation is dose redistribution. The reported scaling is by mask area: a **10×10** mask reduces local dose rate by approximately **100×** compared with a 1-pixel focused beam of the same total flux, and a **32×32** mask yields approximately **1024×** reduction [2410.18348]. Demonstrations include imaging of **sub-pixel pinholes** of **5–20 \(\mu\)m** under very low SNR, as well as binary and gray test objects including patterned Ta and aluminium foils; reconstructed images are said to match ground truth with **mean absolute errors of approximately 0.1 gray level per pixel** [2410.18348].

The limitations are equally explicit. The method requires **over-scan beyond the object edges by at least the mask footprint**, which increases acquisition time. Mask fabrication becomes more complex as the number of gray levels increases and the pixel size decreases. In the reported beamline experiments, **stage and data-transfer overheads limited scan speed to approximately 3 s per point**, preventing large-area scans [2410.18348].

The paper also suggests several extensions: application to **X-ray fluorescence** and other scanning modalities, **static coded-aperture imaging** with large Huffman masks, **3D Huffman arrays** such as **11×11×11** arrays with \(|H|\le3\), and **adaptive mask design via hybrid reverse-Monte-Carlo** optimization [2410.18348]. A plausible implication is that HiXray, in this second sense, should be understood less as a conventional focusing architecture than as a structured-illumination and inverse-problem framework for trading local energy concentration against computational reconstruction fidelity.

Source: https://www.emergentmind.com/topics/hixray