A method for suppressing folded clutter based on range gating and alternating inversion

Abstract

The present invention provides a method for suppressing folded clutter based on range gating and alternating inversion. The method estimates the range distribution interval of clutter echoes from an airborne radar based on the airborne radar scene; determines the spatial-Doppler distribution area of the clutter echoes according to the airborne radar platform parameters and the range distribution interval; determines the maximum number of range segments N for the clutter echo distribution according to the airborne radar parameters, with both the number of each range segment and its clutter echo being n; reconstructs the clutter echoes of the nth range segment using a clutter echo reconstruction submodule; and processes the reconstructed result of the clutter echoes of the range segments other than the nth range segment using a receive filter bank for the nth range segment to obtain the clutter of the nth range segment. The above steps are repeated until iterative convergence is achieved. For each range segment, folded clutter is suppressed by subtracting the reconstructed clutter outside the range segment from the total echo. The method has the advantages of low computational complexity, low complexity, and good robustness.

Description

Technical Field

[0001] The invention belongs to the technical field of airborne radar signal processing, and in particular relates to a method for suppressing folded clutter based on range gating and alternating inversion. Background Art

[0002] In airborne radar applications, strong clutter echoes from the ground or sea surface severely impact radar detection performance of aerial targets. For clutter suppression in airborne radars, narrowband pulse-Doppler (PD) radar systems are primarily used in China. Using space-time adaptive processing (STAP) technology based on phased array antennas, the clutter covariance matrix is estimated from radar echo data to calculate the optimal weighting vector to remove the effects of clutter. However, current STAP technology suffers from high computational complexity, poor suppression of non-uniform clutter, and inability to effectively address the folded clutter problem in range ambiguity. Because parameters such as the pulse repetition time (PRT), carrier frequency, and intra-pulse modulation are identical for each pulse in traditional airborne PD radars, radar echoes exhibit a certain periodicity along the range and velocity dimensions, resulting in range and velocity ambiguity. Typically, airborne PD radars use medium or high pulse repetition frequencies to achieve a larger unambiguous velocity measurement range, thereby creating a larger "clutter-free zone" in the imaging plane for effective separation of clutter and targets. However, medium and high pulse repetition frequencies also introduce significant range ambiguity. For ease of description, the range interval corresponding to the echo delay within the PRT range is defined as a unit range segment. From the perspective of clutter echoes, clutter echoes from different range segments received by the radar antenna's main and side lobes are folded and added in the time domain. The energy of this folded clutter is far greater than the echo energy of long-range aerial targets, severely affecting the radar's target detection performance. Summary of the Invention

[0003] The present invention overcomes one of the deficiencies of the prior art and provides a method for suppressing folded clutter based on range gating and alternating inversion. The method can decorrelate clutter between different range segments and complete blind reconstruction of clutter in different range segments. The method has the advantages of low computational complexity, no need for prior statistical information of clutter, and good robustness.

[0004] According to one aspect of the present disclosure, a method for suppressing folded clutter based on range gating and alternating inversion is proposed, the method comprising:

[0005] Step S1: estimating the distance distribution interval of the clutter echo of the airborne radar based on the airborne radar scene;

[0006] Step S2: determining the airspace-Doppler distribution area of the clutter echo according to the airborne radar platform parameters and the distance distribution interval;

[0007] Step S3: determining the maximum distance segment number N of the clutter echo distribution according to the airborne radar parameters, wherein the number of each distance segment corresponding to the maximum distance segment number N and the number of the clutter echo are both n, and the parameters n and N are positive integers, and 1≤n≤N;

[0008] Step S4: for the nth distance segment, reconstruct the clutter echoes of the distance segments other than the nth distance segment using the clutter echo reconstruction submodule to obtain the clutter echo reconstruction results of the distance segments other than the nth distance segment;

[0009] Step S5: using the receive filter bank of the nth distance segment, processing the result of reconstructing the clutter echoes of other distance segments without the nth distance segment from the total echo to obtain the clutter of the nth distance segment;

[0010] Step S6: n is incremented by 1. When n>N, n=1, and steps S4 to S7 are repeated until the difference between two adjacent iteration results is less than a preset threshold.

[0011] Step S7: For each range segment, the reconstructed clutter not in the range segment is subtracted from the total echo to complete the suppression of the folded clutter.

[0012] In a possible implementation, the distance distribution interval of the clutter echo [R min ,R max ] is determined by the height H of the airborne radar platform p Decide;

[0013] The distance distribution interval of the clutter echo is:

[0014] In a possible implementation, the airborne radar platform parameters include: airborne radar platform movement speed V p The azimuth angle between the main lobe of the airborne radar transmission beam and the direction of movement of the airborne radar platform is θ, and the azimuth angle of the main lobe of the transmission beam close to the direction of movement of the airborne radar platform is θ. n The main lobe and side lobe of the transmitting beam on the side away from the direction of movement of the airborne radar platform cover the azimuth angle of space θ f .

[0015] In a possible implementation, determining the airspace-Doppler distribution area of the clutter echo according to the airborne radar platform parameters and the distance distribution interval of the clutter echo includes:

[0016] Calculating the velocity distribution interval of the clutter echo according to the airborne radar platform parameters;

[0017] Obtaining the Doppler distribution interval of the clutter echo according to the velocity distribution interval of the clutter echo and a Doppler calculation formula;

[0018] The spatial-Doppler distribution area of the clutter echo is obtained according to the time distribution interval and the velocity distribution interval of the clutter echo.

[0019] In a possible implementation, the maximum number of distance segments N of the clutter echo distribution is:

[0020]

[0021] Where C is the speed of light, To round up, PRT is the pulse repetition period.

[0022] In one possible implementation, the clutter echo reconstruction submodule includes five processes: fast-time pulse compression processing, slow-time weighted FFT processing, filtering out energy outside the range-Doppler plane clutter distribution interval, slow-time weighted FFT inverse processing, and fast-time pulse compression inverse processing.

[0023] In a possible implementation, the velocity distribution interval of the clutter echo is:

[0024]

[0025] The Doppler distribution interval of the clutter echo is:

[0026]

[0027] Where λ is the wavelength.

[0028] In a possible implementation, the spatial-Doppler distribution area of the clutter echo is:

[0029]

[0030] in, is the normalized Doppler frequency, ψ is the cone angle variable between the clutter scatterer and the radar array, and f dm is the maximum Doppler frequency of the clutter, α is the yaw angle of the radar array, is the radar array elevation angle, R e Represents the radius of curvature of the Earth.

[0031] The present invention's folded clutter suppression method based on range gating and alternating inversion utilizes a pulse-to-pulse agile waveform with range gating characteristics. When reconstructing clutter in each range segment, it suppresses clutter energy outside the specified range segment through an alternating inversion process, achieving decorrelation between clutter in different range segments and thus completing blind reconstruction of range-segmented clutter. This method has the advantages of low computational complexity, no need for prior clutter statistical information, and good robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings are used to provide a further understanding of the technical solution of this application or the prior art and constitute a part of the specification. Among them, the drawings that express the embodiments of this application are used together with the embodiments of this application to explain the technical solution of this application, but do not constitute a limitation of the technical solution of this application.

[0033] Figure 1 A schematic diagram of a clutter echo and reception filter bank at different distances based on a pulse-to-pulse agility waveform according to an embodiment of the present disclosure is shown;

[0034] Figure 2 A flow chart of a method for suppressing folded clutter based on range gating and alternating inversion according to an embodiment of the present disclosure is shown;

[0035] Figure 3 A flow chart of a method for suppressing folded clutter based on range gating and alternating inversion according to another embodiment of the present disclosure is shown;

[0036] Figure 4a A schematic diagram of the PD processing results of the simulation point clutter echo data based on the traditional clutter suppression method is shown;

[0037] Figure 4b A schematic diagram of the processing results of simulated point clutter echo data PD according to a method for suppressing folded clutter based on range gating and alternating inversion according to an embodiment of the present disclosure is shown;

[0038] Figure 5a The figure shows the result of PD processing of simulated target plus surface clutter echo data based on traditional linear frequency modulation transmission waveform;

[0039] Figure 5b FIG2 shows a schematic diagram of PD processing results of simulated target plus surface clutter echo data based on pulse-to-pulse agile transmission waveform according to an embodiment of the present disclosure;

[0040] Figure 6a The figure shows the result of PD processing of measured echo data based on the traditional linear frequency modulation transmission waveform;

[0041] Figure 6b A schematic diagram of PD processing results of measured echo data based on a pulse-to-pulse agile transmission waveform according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0042] The following will describe the implementation methods of the present invention in detail with reference to the accompanying drawings and examples, so that the implementation process of how the present invention applies technical means to solve technical problems and achieve corresponding technical effects can be fully understood and implemented accordingly. The embodiments of this application and the various features in the embodiments can be combined with each other without conflict, and the technical solutions formed are all within the scope of protection of the present invention.

[0043] In addition, the steps shown in the flowchart of the accompanying drawings can be executed in a computer such as a set of computer-executable instructions. Also, although a logical order is shown in the flowchart, in some cases, the steps shown or described can be performed in a different order than here.

[0044] Figure 1 A schematic diagram of a reception filter bank for clutter echoes and different distance segments based on a pulse-to-pulse agility waveform according to an embodiment of the present disclosure is shown.

[0045] like Figure 1 As shown in the figure, compared with the traditional narrow-band PD radar in which each pulse has the same parameters and modulation, the radar transmission sequence based on the pulse-to-pulse agile waveform has the characteristics of range gating, and the coherent processing of the echoes in the range of interest can be achieved by setting different receiving filter groups.

[0046] Based on the characteristics of clutter echoes from airborne radars, clutter energy at each range is primarily concentrated in the low-speed region of the imaging plane, with the mainlobe and sidelobe clutter echo energy distribution within a velocity range that does not exceed the radar platform velocity. Incorporating the spatial degrees of freedom of phased array radars reveals that clutter energy at each range exhibits a regular elliptical distribution in the airspace-Doppler imaging plane. In particular, if the radar array yaw angle is zero, i.e., when viewing the radar sideways, the elliptical distribution degenerates into a linear one.

[0047] Furthermore, after the receive filter bank is set, clutter echoes from other range segments are mismatched with the receive filter bank, causing their energy to spread across the entire imaging plane. Based on the differences in energy distribution of clutter echoes from different range segments in the imaging plane, the present invention utilizes the range-gating characteristics of the inter-pulse agile waveform to suppress clutter energy outside the target range segment through an alternating inversion process, achieving decorrelation between clutter echoes from different range segments and thus completing blind reconstruction of clutter within each range segment.

[0048] Figure 2 and Figure 3 Flowcharts of a method for suppressing folded clutter based on range gating and alternating inversion according to an embodiment of the present disclosure are shown respectively. Figure 2 and Figure 3 As shown, the method may include:

[0049] Step S1: estimating the distance distribution interval of the clutter echo of the airborne radar based on the airborne radar scene.

[0050] In the airborne PD radar application scenario, assuming that the airborne radar transmission power is large enough, the distance distribution interval of the clutter echo [R min ,R max ]Depends on the flight altitude of the airborne platform p , the specific relationship is shown in formula (1):

[0051]

[0052] Among them, R min is the minimum distance of clutter echo distribution, R max is the maximum distance of clutter echo distribution.

[0053] Step S2: determining the airspace-Doppler distribution area of the clutter echo according to the airborne radar platform parameters and the distance distribution interval.

[0054] Among them, the airborne radar platform parameters include: airborne radar platform movement speed V p The azimuth angle between the main lobe of the airborne radar transmission beam and the direction of movement of the airborne radar platform is θ, and the azimuth angle of the main lobe of the transmission beam close to the direction of movement of the airborne radar platform is θ. n The main lobe and side lobe of the transmitting beam on the side away from the direction of movement of the airborne radar platform cover the azimuth angle of space θ f .

[0055] In one example, determining the airspace-Doppler distribution area of the clutter echo according to the airborne radar platform parameters and the distance distribution interval of the clutter echo may include:

[0056] The velocity distribution interval of the clutter echo is calculated according to the airborne radar platform parameters;

[0057] Obtaining the Doppler distribution interval of the clutter echo according to the velocity distribution interval of the clutter echo and the Doppler calculation formula;

[0058] The spatial-Doppler distribution area of the clutter echo is obtained according to the time distribution interval and the velocity distribution interval of the clutter echo.

[0059] Among them, the velocity distribution range of the clutter echo is:

[0060]

[0061] According to the Doppler calculation formula, the Doppler distribution range of the clutter echo can be calculated as:

[0062]

[0063] Where λ is the wavelength.

[0064] For phased array radar, the distribution of clutter echo in the airspace-Doppler imaging plane It can be obtained from formula (4),

[0065]

[0066] in,

[0067] is the normalized Doppler frequency, ψ is the cone angle variable between the clutter scatterer and the radar array, and f dm is the maximum Doppler frequency of the clutter, α is the yaw angle of the radar array, is the radar array elevation angle, R e Represents the radius of curvature of the Earth.

[0068] In particular, when the yaw angle α = 0, that is, the case of positive side view, the distribution of clutter echo in the airspace-Doppler imaging plane is Degenerated into a linear distribution, the corresponding relationship is:

[0069]

[0070] Step S3: Determine the maximum distance segment number N of the clutter echo distribution based on the airborne radar parameters. The number of each distance segment and its clutter echo number corresponding to the maximum distance segment number N are both n. The parameters n and N are positive integers, and 1≤n≤N. The maximum distance segment number N of the clutter echo distribution is:

[0071]

[0072] Where C is the speed of light, To round up, PRT is the pulse repetition period.

[0073] Step S4: for the nth distance segment, the clutter echo reconstruction submodule is used to reconstruct the clutter echoes of the distance segments other than the nth distance segment, and obtain the clutter echo reconstruction results of the distance segments other than the nth distance segment.

[0074] Step S5: using the reception filter bank of the nth distance segment, processing the total echo minus the clutter echo reconstruction result of other distance segments excluding the nth distance segment to obtain the clutter of the nth distance segment.

[0075] Step S6: n is incremented by 1. When n>N, n=1, and steps S4 to S7 are repeated until the difference between two adjacent iteration results is less than a preset threshold.

[0076] like Figure 1 As shown in FIG, a receive filter group is set for the nth range segment. If the parameters and modulation form within each pulse of the inter-pulse agile waveform are the same and only the initial phase between pulses is different, the receive filter group can be designed using a delay-matched filter group. If the modulation form within each pulse is different, the receive filter group must use a joint mismatched filter to improve the range sidelobe modulation (RSM) effect caused by the change in the inter-pulse modulation form. The above can be achieved using several commonly used joint mismatched filter design algorithms, which will not be introduced in detail here.

[0077] The total echo is used to subtract clutter from all other range segments outside the nth range segment that have already been processed by the clutter reconstruction submodule. The result of this subtraction is then processed by the clutter reconstruction submodule using the receive filter bank for the nth range segment, yielding the clutter for the nth range segment processed by the clutter reconstruction submodule. The clutter reconstruction submodule includes five steps: fast-time pulse compression, slow-time weighted FFT processing, filtering out energy outside the range-Doppler plane clutter distribution interval, inverse slow-time weighted FFT processing, and inverse fast-time pulse compression processing. If the radar is a phased array, the spatial degrees of freedom of the radar array can be exploited. A more refined energy filtering process can be performed in the spatial-Doppler imaging plane by combining the spatial processing module. The clutter reconstruction submodule includes three steps: spatial-weighted FFT processing, filtering out energy outside the spatial-Doppler plane clutter distribution interval, and inverse spatial-weighted FFT processing.

[0078] Increment n by 1 (if n>N after the increment, reset n=1) and repeat steps S4-S7 until the iterative process converges. The iterative convergence condition can be determined by the difference between the results of two consecutive iterations. If the absolute value of the difference is less than a certain threshold, the algorithm is considered to have converged.

[0079] Step S7: For each range segment, the reconstructed clutter not in the range segment is subtracted from the total echo to complete the suppression of the folded clutter.

[0080] Figure 4a A schematic diagram of the PD processing results of the simulation point clutter echo data based on the traditional clutter suppression method is shown; Figure 4b A schematic diagram of the processing results of simulated point clutter echo data PD according to a method for suppressing folded clutter based on range gating and alternating inversion according to an embodiment of the present disclosure is shown.

[0081] In order to test the performance of the range-gated folded clutter suppression method, a simulation test is carried out on the cases where point clutter exists in the first two range segments based on the inter-pulse initial phase agility waveform.

[0082] Assume that the airborne radar transmits pulses in the form of linear frequency modulation, the initial phase between pulses is generated by a random phase uniformly distributed between, the radar unambiguous ranging range is 75km, and the speed of the two point clutter is 0. When point clutter exists in the first two range segments and the folded clutter blind reconstruction process is not performed, the output PD result is as follows: Figure 4a As shown. Figure 4a It can be seen that the two point clutters have a higher energy spread in the distance segment where the other is located. This is because the receiving filter group is mismatched for the echoes in other distance segments. When the echo is a surface clutter or volume clutter distributed in a certain distance range, the energy of the folded clutter will be spread over the entire imaging plane, seriously affecting the radar target detection performance. Using the folded clutter blind reconstruction method proposed in this invention, the PD processing results are as follows: Figure 4b As shown by Figure 4b It can be seen that the point clutter in the two range segments can be effectively reconstructed.

[0083] Figure 5a The figure shows the result of PD processing of simulated target plus surface clutter echo data based on traditional linear frequency modulation transmission waveform; Figure 5b A schematic diagram of PD processing results of simulated target plus surface clutter echo data based on pulse-to-pulse agile transmission waveform according to an embodiment of the present disclosure is shown.

[0084] For example, a comprehensive simulation test of a surface clutter scenario was conducted. In the simulation, the radar's unambiguous ranging range was set to 75km. The distances and speeds of two point targets were 275km, 150m / s, and 325km, -60m / s, respectively, which are located in the 4th and 5th distance segments, respectively. The positive and negative signs of the speed represent that the target is moving away from or toward the radar, respectively. The target signal-to-noise ratio is about 20dB. The speed distribution range of the simulated surface clutter is about -80m / s to 80m / s, the distance distribution range is about 8km to 210km, and the signal-to-noise ratio is about -50dB.

[0085] like Figure 5a As shown in , the clutter and the target are folded into the same range segment, the range ambiguity phenomenon is serious, and since the low-speed point target is covered by the folded clutter, it cannot be effectively detected; Figure 5b As shown in the figure, the folded clutter in each range segment is completely reconstructed, which effectively removes the influence of the folded clutter on the target detection in the 4th and 5th range segments, and realizes the unambiguous ranging and speed measurement of the target.

[0086] Figure 6a The figure shows the result of PD processing of measured echo data based on the traditional linear frequency modulation transmission waveform; Figure 6b A schematic diagram of PD processing results of measured echo data based on a pulse-to-pulse agile transmission waveform according to an embodiment of the present disclosure is shown.

[0087] To further prove the effectiveness of the method, the present invention was also verified on measured data. The test process was carried out on the ground, and the traditional linear frequency modulation signal and the pulse-to-pulse agile waveform were transmitted alternately in the test to compare the performance of the traditional PD processing method and the method of the present invention under the same scenario. Figure 6a It can be seen that when the radar transmits a traditional linear frequency modulation signal, the clutter and target echo are folded into the same fuzzy range segment, resulting in serious range ambiguity. Figure 6b As shown, by adopting the pulse-to-pulse agile waveform and processing it through the method of the present invention, clutter echoes from different distance segments can be effectively reconstructed, and folded clutter can be effectively suppressed, so that the target can be effectively detected.

[0088] The present invention provides a method for suppressing folded clutter based on range gating and alternating inversion, comprising the steps of: estimating the range distribution interval of the clutter echo of the airborne radar based on the airborne radar scene; determining the spatial-Doppler distribution area of the clutter echo according to the airborne radar platform parameters and the range distribution interval; determining the maximum range segment number N of the clutter echo distribution according to the airborne radar parameters, wherein the number of each range segment corresponding to the maximum range segment number N and the number of the clutter echo thereof are both n, and the parameters n and N are positive integers, and 1≤n≤N; and reconstructing the clutter echo sub-area for the nth range segment using the clutter echo sub-area. The module reconstructs the clutter echoes of the range segments other than the nth range segment, obtaining a reconstructed result of the clutter echoes of the range segments other than the nth range segment. Step S5: Using the receive filter bank for the nth range segment, the total echo minus the reconstructed result of the clutter echoes of the range segments other than the nth range segment is processed to obtain the clutter of the nth range segment. Step S6: n is incremented by 1, and steps 4 through S7 are repeated until the difference between the results of two consecutive iterations is less than a preset threshold. Step S7: For each range segment, the reconstructed clutter outside the range segment is subtracted from the total echo to suppress the folded clutter. When reconstructing the clutter of each range segment, an alternating inversion process is used to suppress the clutter energy outside the range segment, achieving decorrelation between clutter in different range segments, thereby completing blind reconstruction of the range-segmented clutter. This method has the advantages of low computational complexity, no need for prior clutter statistical information, and good robustness.

[0089] Although the embodiments disclosed herein are as described above, the contents described herein are merely embodiments for facilitating understanding of the present invention and are not intended to limit the present invention. Any person skilled in the art may make any modifications and variations in the form and details of the embodiments without departing from the spirit and scope of the present invention. However, the scope of patent protection of the present invention shall remain subject to the scope defined by the appended claims.

Claims

1. A method for suppressing folded clutter based on range gating and alternating inversion, characterized in that: The method comprises: Step S1: estimating the distance distribution interval of the clutter echo of the airborne radar based on the airborne radar scene; Step S2: determining the airspace-Doppler distribution area of the clutter echo according to the airborne radar platform parameters and the distance distribution interval; Step S3: determining the maximum distance segment number N of the clutter echo distribution according to the airborne radar parameters, wherein the number of each distance segment corresponding to the maximum distance segment number N and the number of the clutter echo are both n, and the parameters n and N are positive integers, and 1≤n≤N; Step S4: for the nth distance segment, reconstruct the clutter echoes of the distance segments other than the nth distance segment using the clutter echo reconstruction submodule to obtain the clutter echo reconstruction results of the distance segments other than the nth distance segment; Step S5: using the receive filter bank of the nth distance segment, processing the result of reconstructing the clutter echoes of other distance segments without the nth distance segment from the total echo to obtain the clutter of the nth distance segment; Step S6: n is incremented by 1. When n>N, n=1, and steps S4 to S6 are repeated until the difference between two adjacent iteration results is less than a preset threshold. Step S7: For each range segment, the reconstructed clutter not in the range segment is subtracted from the total echo to complete the suppression of the folded clutter.

2. The suppression method according to claim 1, characterized in that The distance distribution interval of the clutter echo [R min ,R max ] is determined by the height H of the airborne radar platform p Decide; The distance distribution interval of the clutter echo is:

3. The suppression method according to claim 1, characterized in that The airborne radar platform parameters include: airborne radar platform movement speed V p The azimuth angle between the main lobe of the airborne radar transmission beam and the direction of movement of the airborne radar platform is θ, and the azimuth angle of the main lobe of the transmission beam close to the direction of movement of the airborne radar platform is θ. n The main lobe and side lobe of the transmitting beam on the side away from the direction of movement of the airborne radar platform cover the azimuth angle of space θ f .

4. The suppression method according to claim 3, characterized in that: The determining of the airspace-Doppler distribution area of the clutter echo according to the airborne radar platform parameters and the distance distribution interval of the clutter echo comprises: Calculating the velocity distribution interval of the clutter echo according to the airborne radar platform parameters; Obtaining the Doppler distribution interval of the clutter echo according to the velocity distribution interval of the clutter echo and a Doppler calculation formula; The spatial-Doppler distribution area of the clutter echo is obtained according to the time distribution interval and the velocity distribution interval of the clutter echo.

5. The suppression method according to claim 3, characterized in that: The maximum distance segment number N of the clutter echo distribution is: Where C is the speed of light, To round up, PRT is the pulse repetition period.

6. The suppression method according to claim 1, characterized in that: The clutter echo reconstruction submodule includes five processes: fast time pulse compression processing, slow time weighted FFT processing, filtering out energy outside the range-Doppler plane clutter distribution interval, slow time weighted FFT inverse processing and fast time pulse compression inverse processing.

7. The suppression method according to claim 4, characterized in that: The velocity distribution interval of the clutter echo is: The Doppler distribution interval of the clutter echo is: Where λ is the wavelength.

8. The suppression method according to claim 7, characterized in that: The spatial-Doppler distribution area of the clutter echo is: in, is the normalized Doppler frequency, ψ is the cone angle variable between the clutter scatterer and the radar array, and f dm is the maximum Doppler frequency of the clutter, α is the yaw angle of the radar array, is the radar array elevation angle, R e Represents the radius of curvature of the Earth.

Images