1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
// Copyright 2018 Stefan Kroboth
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// http://apache.org/licenses/LICENSE-2.0> or the MIT license <LICENSE-MIT or
// http://opensource.org/licenses/MIT>, at your option. This file may not be
// copied, modified, or distributed except according to those terms.

//! # Shepp-Logan phantom
//!
//! Have you ever had the need to create hundreds to thousands of Shepp-Logan phantoms per second?
//! Well if you do, you're doing something wrong, but you've come to the right place.
//! The Shepp-Logan phantom is a numerical phantom which is defined as the sum of 10 ellipses. It
//! is often used as a test image for image reconstruction algorithms.
//! This crate provides a dependency-free, efficient implementation for creating Shepp-Logan
//! phantoms in 2D.
//! The following results were obtained with `cargo bench` on an Intel Core i7 with 2.70GHz:
//!
//! Resolution | time        | fps   
//! -----------|-------------|------
//! 128x128    |   111,000ns | 9000  
//! 256x256    |   440,000ns | 2200  
//! 512x512    | 1,780,000ns |  560  
//!
//! Two versions are provided: The original version as described in [0] and a modified version,
//! which has higher contrast as described in [1]. If you do not know the difference between those
//! two, you most likely want the modified version.
//!
//! To use the crate, add `shepplogan` to your `Cargo.toml`:
//!
//! ```toml
//! shepplogan = "^1"
//! ```
//!
//! # Example
//!
//! ```rust
//! extern crate shepplogan;
//! use shepplogan::{shepplogan, shepplogan_modified};
//!
//! // Dimensions of the image grid
//! let (nx, ny) = (256, 320);
//!
//! // Original Shepp-Logan Phantom (the dynamic range is between 0.0 and 2.0)
//! let phantom = shepplogan(nx, ny);
//!
//! // Modified Shepp-Logan Phantom (the dynamic range is between 0.0 and 1.0)
//! let phantom_modified = shepplogan_modified(nx, ny);
//! ```
//!
//! See `examples/example.rs` for an example which saves the phantom to disk.
//!
//! You can also create your own phantom by defining ellipses:
//!
//! ```rust
//! extern crate shepplogan;
//! use shepplogan::{phantom, Ellipse};
//!
//! // Dimensions of the image grid
//! let (nx, ny) = (256, 320);
//!
//! // Define two ellipses
//! let ellipses =
//!     [
//!         Ellipse::new(0.0, -0.0184, 0.6624, 0.874, 0.0, -0.98),
//!         Ellipse::new(0.0, 0.0, 0.69, 0.92, 0.0, 2.0),
//!     ];
//!
//! let ph = phantom(&ellipses, nx, ny);
//! ```
//!
//! This will create a phantom consisting of two ellipses.
//!
//! # References
//!
//! [0] Shepp, LA and Logan BF, "The Fourier reconstruction of a head section." IEEE Transactions
//! on Nuclear Science 21, No. 3 (1974)
//!
//! [1] Toft, PA, "The Radon Transform - Theory and Implementation", PhD dissertation, Departement
//! of Mathematical Modelling, Technical University of Denmark (1996)

#![warn(missing_docs)]
#[cfg(feature = "parallel")]
extern crate rayon;

mod ellipse;
pub use crate::ellipse::Ellipse;
#[cfg(feature = "parallel")]
use rayon::prelude::*;
#[cfg(feature = "parallel")]
use std::sync::Mutex;

macro_rules! parts {
    () => {
        [
            Ellipse::new(0.0, 0.35, 0.21, 0.25, 0.0, 0.01),
            Ellipse::new(0.0, 0.1, 0.046, 0.046, 0.0, 0.01),
            Ellipse::new(0.0, -0.1, 0.046, 0.046, 0.0, 0.01),
            Ellipse::new(-0.08, -0.605, 0.046, 0.023, 0.0, 0.01),
            Ellipse::new(0.0, -0.605, 0.023, 0.023, 0.0, 0.01),
            Ellipse::new(0.06, -0.605, 0.023, 0.046, 0.0, 0.01),
            Ellipse::new(0.22, 0.0, 0.11, 0.31, -18.0, -0.02),
            Ellipse::new(-0.22, 0.0, 0.16, 0.41, 18.0, -0.02),
            Ellipse::new(0.0, -0.0184, 0.6624, 0.874, 0.0, -0.98),
            Ellipse::new(0.0, 0.0, 0.69, 0.92, 0.0, 2.0),
        ]
    };
}

macro_rules! parts_modified {
    () => {
        [
            Ellipse::new(0.0, 0.35, 0.21, 0.25, 0.0, 0.1),
            Ellipse::new(0.0, 0.1, 0.046, 0.046, 0.0, 0.1),
            Ellipse::new(0.0, -0.1, 0.046, 0.046, 0.0, 0.1),
            Ellipse::new(-0.08, -0.605, 0.046, 0.023, 0.0, 0.1),
            Ellipse::new(0.0, -0.605, 0.023, 0.023, 0.0, 0.1),
            Ellipse::new(0.06, -0.605, 0.023, 0.046, 0.0, 0.1),
            Ellipse::new(0.22, 0.0, 0.11, 0.31, -18.0, -0.2),
            Ellipse::new(-0.22, 0.0, 0.16, 0.41, 18.0, -0.2),
            Ellipse::new(0.0, -0.0184, 0.6624, 0.874, 0.0, -0.8),
            Ellipse::new(0.0, 0.0, 0.69, 0.92, 0.0, 1.0),
        ]
    };
}

/// Original Shepp-Logan phantom
///
/// Constructs the original Shepp-Logan phantom as described in:
///
/// Shepp, LA and Logan BF, "The Fourier reconstruction of a head section." IEEE Transactions on
/// Nuclear Science 21, No. 3 (1974)
///
/// The parameters `nx` and `ny` define the number of pixels in `x` and `y` direction.
/// The dynamic range of the values is between `0.0` and `2.0`.
pub fn shepplogan(nx: usize, ny: usize) -> Vec<f64> {
    let ellipses = parts!();
    phantom(&ellipses, nx, ny)
}

/// Modified Shepp-Logan phantom with better contrast
///
/// Constructs the modified Shepp-Logan phantom as described in:
///
/// Toft, PA, "The Radon Transform - Theory and Implementation", PhD dissertation, Departement of
/// Mathematical Modelling, Technical University of Denmark (1996)
///
/// The parameters `nx` and `ny` define the number of pixels in `x` and `y` direction.
/// The dynamic range of the values is between `0.0` and `1.0`.
pub fn shepplogan_modified(nx: usize, ny: usize) -> Vec<f64> {
    let ellipses = parts_modified!();
    phantom(&ellipses, nx, ny)
}

#[cfg(not(feature = "parallel"))]
#[cfg(not(feature = "slow_impl"))]
/// Creates a phantom based on given ellipses
///
/// Besides `nx` and `ny`, which define the number of pixels in `x` and `y` direction, this
/// function also requires a vector of Ellipses.
pub fn phantom(ellipses: &[Ellipse], nx: usize, ny: usize) -> Vec<f64> {
    let mut arr = vec![0.0; nx * ny];
    let nx2 = (nx as f64) / 2.0;
    let ny2 = (ny as f64) / 2.0;
    let nmin = (std::cmp::min(nx, ny) as f64) / 2.0;

    for e in ellipses.iter() {
        let bbox = e.bounding_box(nx, ny);
        for x in bbox.0..bbox.2 {
            let xi = (x as f64 - nx2) / nmin;
            for y in bbox.1..bbox.3 {
                let yi = (y as f64 - ny2) / nmin;
                if e.inside(xi, yi) {
                    arr[(ny - y) * nx + x] += e.intensity();
                }
            }
        }
    }
    arr
}

#[cfg(feature = "parallel")]
/// Creates a phantom based on given ellipses
///
/// Besides `nx` and `ny`, which define the number of pixels in `x` and `y` direction, this
/// function also requires a vector of Ellipses.
pub fn phantom(ellipses: &[Ellipse], nx: usize, ny: usize) -> Vec<f64> {
    let arr: Vec<Mutex<f64>> = (0..(nx * ny))
        .into_par_iter()
        .map(|_| Mutex::new(0.0))
        .collect();
    let nx2 = (nx as f64) / 2.0;
    let ny2 = (ny as f64) / 2.0;
    let nmin = (std::cmp::min(nx, ny) as f64) / 2.0;

    ellipses.into_par_iter().for_each(|e| {
        let bbox = e.bounding_box(nx, ny);
        (bbox.0..bbox.2).into_iter().for_each(|x| {
            let xi = (x as f64 - nx2) / nmin;
            (bbox.1..bbox.3).into_iter().for_each(|y| {
                let yi = (y as f64 - ny2) / nmin;
                if e.inside(xi, yi) {
                    let mut b = arr[(ny - y) * nx + x].lock().unwrap();
                    *b = *b + e.intensity();
                }
            })
        });
    });
    arr.into_par_iter().map(|x| *(x.lock().unwrap())).collect()
}

#[cfg(feature = "slow_impl")]
/// Creates a phantom based on given ellipses
///
/// Besides `nx` and `ny`, which define the number of pixels in `x` and `y` direction, this
/// function also requires a vector of Ellipses.
pub fn phantom(ellipses: &[Ellipse], nx: usize, ny: usize) -> Vec<f64> {
    let mut arr = Vec::with_capacity(nx * ny);
    let nx2 = (nx as f64) / 2.0;
    let ny2 = (ny as f64) / 2.0;
    let nmin = (std::cmp::min(nx, ny) as f64) / 2.0;

    for y in 0..ny {
        for x in 0..nx {
            let xi = (x as f64 - nx2) / nmin;
            let yi = ((ny - y) as f64 - ny2) / nmin;
            arr.push(
                ellipses
                    .iter()
                    .filter(|e| e.inside(xi, yi))
                    .map(|e| e.intensity())
                    .sum(),
            );
        }
    }
    arr
}