A method for identifying and labeling a maximum non-blind-element square region of an area array detector

Abstract

This invention discloses a method for identifying and labeling the largest blind-element-free square region in an area array detector. The method includes the following steps: 1. Initializing the blind element data matrix and the result data matrix; 2. Calculating the side length of the largest blind-element-free square that can be expanded downwards and to the right from each element in the blind element data matrix, and storing the result in the result data matrix; 3. Calculating the maximum value and its corresponding coordinates in the result data matrix, which represent the size of the desired largest blind-element-free square region and the coordinates of its top-left vertex; 4. Drawing the blind element distribution map of the original blind element matrix; 5. Drawing each largest blind-element-free square region on the blind element distribution map and labeling its size. The advantage of this invention is that it achieves automatic identification and labeling of the largest blind-element-free square region in an area array detector, and the method is simple.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of area array detector, in particular to a kind of area array detector blind element analysis method. BACKGROUND

[0002] Area array detector will inevitably introduce blind element due to material defects and chip process. With the development of imaging technology, the application scenarios of area array detector are more and more extensive. In addition to the blind element rate, the aggregation of blind element also begins to attract more and more attention of researchers. In some application scenarios that require accurate tracking of target position, although the blind element rate is high, the blind element is relatively concentrated in some areas, so that the detector with large blind element-free area performs better than the detector with low blind element rate and dispersed blind element distribution. Some projects have also taken "maximum blind element-free area ≥ n × n" as an important index for investigating the detector. Therefore, we not only need to identify the position of blind element, but also need to identify the blind element-free area of the detector. The existing technical methods are specific analysis methods for blind element, such as the method for counting consecutive blind elements mentioned in the article "Statistical Analysis of Infrared Focal Plane Detector Blind Element by MATLAB" (Infrared, Vol. 30, No. 3, pp. 43-45) by Mao Jingxiang et al. However, there is no identification method for blind element-free area. SUMMARY

[0003] The present application aims to provide a method for identifying and labeling the maximum blind element-free square area of area array detector. The method can identify the maximum blind element-free square area on the blind element distribution map of area array detector and label it on the blind element distribution map, while counting the number of such areas.

[0004] To achieve the above-mentioned purpose, the present application provides a method for identifying and labeling the maximum blind element-free square area of area array detector, which comprises the following steps:

[0005] Step 1: Initialize blind element data matrix BP and result data matrix Result, where blind element data matrix BP = [bp ij ], the size of matrix BP is the size of area array detector, and the size is I × J, and bp

[0006]

[0007] In the formula, i and j are subscripts, i takes the value range of 1 to I, j takes the value range of 1 to J, result data matrix Result = [r ij ], the size of matrix Result is the size of area array detector, and the size is I × J, and r ij represents the element bp ijStarting from r, the side length of the largest unblinded element square that can be expanded downwards and to the right, where i and j are subscripts, i ranging from 1 to I, j ranging from 1 to J, and r ij Position and bp ij One-to-one correspondence, initialize the matrix Result as a matrix of all zeros;

[0008] Step 2: Calculate r ij Store the result in the matrix Result, where r ij The steps to obtain are as follows:

[0009] 2-1) Judgment and r ij The corresponding bp ij If the value is 1, then we get r. ij =0; if it is 0, proceed to the next step.

[0010] 2-2) From bp ij Start by expanding the square region to the lower right. For each increase of 1 in the square's side length, calculate the sum of all elements within that region and check if the sum is 0. If it is 0, and the square's boundary has not exceeded the area matrix boundary, continue expanding the square. If the sum is not 0, or the square's boundary has exceeded the area matrix boundary, record the current side length 'a' of the square, thus obtaining 'r'. ij = a–1;

[0011] 2-3) Repeat steps 2-1) and 2-2), traversing each element bp in matrix BP starting from the top left corner. ij Each bp is calculated ij The corresponding r ij The value;

[0012] Step 3: Calculate the maximum value Max_R of the matrix Result and its corresponding coordinates. This represents the size of the largest blind-free square region and the coordinates of its top-left vertex. Assuming there are n such regions, we obtain n pairs of position coordinates.

[0013] Max_R = max(max(Result)) (2)

[0014] [jm, im] = find(Result == Max_R) (3)

[0015] n = length(jm) (4)

[0016] Where `max`, `find`, and `length` are functions in MATLAB, the resulting [jm, im] is an n x 2 matrix, each row containing a pair of position coordinates, with the first column representing the ordinate and the second column representing the abscissa, in the form of...

[0017]

[0018] Where jm_k is the ordinate of the top left corner of the kth largest unblinded element square region, and im_k is the abscissa of the top left corner of the kth largest unblinded element square region;

[0019] Step 4: Draw the blind element distribution diagram of the original blind data matrix BP.

[0020] Imshow(~BP)(6)

[0021] hold on (7)

[0022] Imshow and hold are functions in the MATLAB software. The purpose of the hold on statement is to keep the plotting interface open in preparation for the subsequent labeling of the largest unblinded square region.

[0023] Step 5: On the blind pixel distribution map obtained in Step 4, draw each maximum blind pixel-free square region. The coordinates of the upper left vertex of the k-th maximum blind pixel-free square region are (im_k, jm_k), and its size is Max_R × Max_R. Also, label the dimensions of the region in the lower right corner of the box.

[0024] x_k = [im_k, im_k + Max_R – 1, im_k + Max_R – 1, im_k, im_k] (8)

[0025] y_k = [jm_k, jm_k, jm_k + Max_R -1, jm_k + Max_R -1, jm_k] (9)

[0026] plot(x_k,y_k,'k')(10)

[0027] text(im_k+Max_R–1+q,jm_k+Max_R–1+q,[num2str(Max_R),'x',num2str(Max_R)],'color','k')(11)

[0028] In the MATLAB software, plot and text are functions. In the plot and text statements, 'k' indicates that the color of the drawn box is black, but other colors can also be set, such as b for blue and r for red. In the text statement, q is the offset of the marker text from the drawn box, which can be set freely.

[0029] The beneficial effects of this invention are: it realizes the automatic identification and labeling of the largest blind element square region in the area array detector, the method is simple, and it is applicable to area array detectors of various array sizes. Attached Figure Description

[0030] Appendix Figure 1 : A schematic diagram of the method of the present invention.

[0031] Appendix Figure 2 : A binary image of the original blind metadata matrix in a specific embodiment of the present invention.

[0032] Appendix Figure 3 The results of identification and annotation of the largest blind-memory square region in a specific embodiment of the present invention. Detailed Implementation

[0033] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments:

[0034] This embodiment identifies and labels the largest blind-element-free square region of a 512×512 area array detector. (See attached image.) Figure 2 This is a binary image of the original blind data matrix of the detector. Write MATLAB code according to the steps in the instruction manual, and use the written MATLAB code to perform calculations to obtain the attached... Figure 3 The results show that the size of the largest blind-free square region is 67×67; the number of the largest blind-free square regions is 4; and the coordinates of the top left corner of the square are (23, 278), (23, 279), (23, 280), (23, 281).

[0035] The specific examples described above are only for illustrating the present invention in detail and are not intended to limit the present invention in any way.

Claims

1. A method for identifying and labeling the largest blind-element-free square region of an area array detector, characterized in that, Includes the following steps: Step 1: Initialize the blind metadata matrix BP and the result data matrix Result, where the blind metadata matrix BP = [bp ij The size of matrix BP is equal to the size of the area array detector, which is I×J. ； In the formula, i and j are subscripts, where i ranges from 1 to I and j ranges from 1 to J. The resulting data matrix Result = [r ij The size of matrix Result is equal to the array size of the area detector, which is I×J, r ij This indicates that the element bp in BP is used. ij Starting from r, the side length of the largest unblinded element square that can be expanded downwards and to the right, where i and j are subscripts, i ranging from 1 to I, j ranging from 1 to J, and r ij Position and bp ij One-to-one correspondence, initialize the matrix Result as a matrix of all zeros; Step 2: Calculate r ij Store the result in the matrix Result, where r ij The steps to obtain are as follows: 2-1) Judgment and r ij The corresponding bp ij If the value is 1, then we get r. ij =0; if it is 0, proceed to the next step. 2-2) From bp ij Start by expanding the square region to the lower right. For each increase of 1 in the square's side length, calculate the sum of all elements within that region and check if the sum is 0. If it is 0, and the square's boundary has not exceeded the area matrix boundary, continue expanding the square. If the sum is not 0, or the square's boundary has exceeded the area matrix boundary, record the current square side length 'a', thus obtaining r. ij = a–1; 2-3) Repeat steps 2-1) and 2-2), traversing each element bp in matrix BP starting from the top left corner. ij Each bp is calculated ij The corresponding r ij The value; Step 3: Calculate the maximum value Max_R of the matrix Result and its corresponding coordinates. This represents the size of the largest blind-free square region and the coordinates of its top-left vertex. Assuming there are n such regions, we obtain n pairs of position coordinates: Max_R=max(max(Result)) (2) [jm,im]=find(Result==Max_R) (3) n=length(jm) (4) Where max, find, and length are functions in MATLAB, the resulting [jm, im] is an n x 2 matrix, each row containing a pair of position coordinates, with the first column representing the ordinate and the second column representing the abscissa, in the following form: ； Where jm_k is the ordinate of the top left corner of the kth largest unblinded element square region, and im_k is the abscissa of the top left corner of the kth largest unblinded element square region; Step 4: Draw the blind element distribution diagram of the original blind data matrix BP. Imshow(~BP)(6)hold on(7) Imshow and hold are functions in the MATLAB software. The purpose of the hold on statement is to keep the plotting interface open in preparation for the subsequent labeling of the largest unblinded square region. Step 5: On the blind pixel distribution map obtained in Step 4, draw each maximum blind pixel-free square region. The coordinates of the upper left vertex of the k-th maximum blind pixel-free square region are (im_k, jm_k), and its size is Max_R × Max_R. Also, label the dimensions of the region in the lower right corner of the box. x_k=[im_k,im_k+Max_R-1,im_k+Max_R–1,im_k,im_k] (8) y_k=[jm_k,jm_k,jm_k+Max_R-1,jm_k+Max_R-1,jm_k] (9) plot(x_k,y_k,'k')(10) text(im_k+Max_R-1+q,jm_k+Max_R-1+q,[num2str(Max_R),'x',num2str(Max_R)],'color','k')(11) In the MATLAB software, plot and text are functions. In the plot and text statements, 'k' indicates that the color of the drawn box is black, but other colors can also be set. b indicates blue, and r indicates red. In the text statement, q is the offset of the marker text from the drawn box, which can be set freely.

Images