US 20020159622 A1 Abstract A method an apparatus for detecting lines in medical images is disclosed, wherein a direction image array and a line image array are formed by filtering a digital image with a single-peaked filter, convolving the resultant array with second order difference operators oriented along the horizontal, vertical, and diagonal axes, and computing the direction image arrays and line image arrays as direct scalar functions of the results of the second order difference operations. Advantageously, line detection based on the use of four line operator functions along the horizontal, vertical, and diagonal directions in accordance with the preferred embodiments actually results in fewer computations than line detection based on the use of three line operator functions. In particular, because of the special symmetries involved, 3×3 second order difference operators may be effectively used. Moreover, the number of computations associated with the second order difference operations may be achieved with simple register shifts, additions, and subtractions, yielding an overall line detection process that is significantly less computationally intensive than prior art algorithms. Also according to a preferred embodiment, computational complexity is reduced by selecting a separable single-peaked filter, and sequentially convolving the digital image with the component kernels of the separable single-peaked filter.
Claims(33) 1. A method for detecting lines in a digital image, comprising the steps of:
filtering said digital image to produce a filtered image array; convolving said filtered image array with a plurality of second order difference operators designed to extract second order directional derivative information from said filtered image array in a predetermined set of directions; processing information resulting from said step of convolving to produce a line image; wherein said predetermined set of directions is selected to correspond to an aspect ratio of said second order difference operators. 2. The method of 3. The method of 4. The method of selecting a single-peaked filter kernel; and convolving said digital mammogram image with said single-peaked filter kernel. 5. The method of 6. The method of 7. The method of convolving said filtered image array with 3×3 second order difference operators designed to extract second order derivative information along the 45 degree and 135 degree directions; and subsequent to said step convolving said filtered image array with 3×3 second order difference operators designed to extract second order derivative information along the 45 degree and 135 degree directions, multiplying the results of said step by a constant correction factor to accommodate for more widely spaced sampling along the diagonals. 8. A method for detecting lines in a digital image, comprising the steps of:
selecting a spatial scale parameter, said spatial scale parameter corresponding to a desired range of line widths for detection; convolving said digital image with a first one dimensional kernel and a second one dimensional kernel to produce a filtered image array, said first one dimensional kernel and said second one dimensional kernel each having a size related to said spatial scale parameter; producing a line image based on second-order spatial derivatives of said filtered image array; wherein said line image is produced from said digital image using a number of computations that is substantially proportional to the spatial scale parameter such that, as the spatial scale parameter is increased, said number of computations increases at a rate that is less than the rate of increase of the square of the spatial scale parameter. 9. The method of convolving said filtered image array with a plurality of second order difference operators designed to extract second order directional derivative information from said filtered image array in a predetermined set of directions; and processing information resulting from said step of convolving to produce a line image; wherein said predetermined set of directions includes directions along the diagonals of the digital mammogram image. 10. The method of 11. The method of 12. The method of 13. A method for detecting lines in a digital image, comprising the steps of:
selecting a spatial scale parameter, said spatial scale parameter corresponding to a desired range of line widths for detection; convolving said digital image with a first one dimensional kernel and a second one dimensional kernel to produce a filtered image array, said first one dimensional kernel and said second one dimensional kernel each having a size related to said spatial scale parameter; separately convolving said filtered image array with a first, second, and third second order difference operator to produce a first, second, and third resulting array, respectively; computing a direction image array comprising, at each pixel, a first predetermined scalar function of corresponding pixel values in said first, second, and third resulting arrays; computing a line intensity function array comprising, at each pixel, a second predetermined scalar function of corresponding pixel values in said first, second, and third resulting arrays; and computing a line image array using information in said line intensity function array. 14. The method of 15. The method of 16. The method of 17. The method of 18. The method of 19. The method of 20. The method of 21. A computer-readable medium which can be used for directing an apparatus to detect lines in a digital image, comprising:
means for directing said apparatus to filter said image to produce a filtered array; means for directing said apparatus to convolve said filtered image array with a plurality of second order difference operators designed to extract second order directional derivative information from said filtered image array in a predetermined set of directions; means for directing said apparatus to process information resulting from said step of convolving to produce a line image; wherein said predetermined set of directions is selected to correspond to an aspect ratio of said second order difference operators. 22. The computer-readable medium of 23. The computer-readable medium of 24. The computer-readable medium of 25. The computer-readable medium of 26. The computer-readable medium of 27. An apparatus for detecting lines in digital images, said apparatus comprising:
a first memory for storing a digital image; a first convolution device capable of convolving said digital image with a first one dimensional kernel and a second one dimensional kernel to produce a filtered image array, said first one dimensional kernel and said second one dimensional kernel each having a size related to the size of lines being detected; a second convolution device capable of separately convolving said filtered image array with a first, a second, and a third second order difference operator to produce a first, second, and third resulting array, respectively; a first processing device capable of computing a direction image array comprising, at each pixel, a first predetermined scalar function of corresponding pixel values in said first, second, and third resulting arrays; a second processing device capable of computing a line intensity function array comprising, at each pixel, a second predetermined scalar function of corresponding pixel values in said first, second, and third resulting arrays; and a third processing device capable of computing a line image array using information in said line intensity function array. 28. The method of 29. The method of 30. The method of 31. The method of 32. The method of 33. The method of Description [0001] The present invention relates to the field of computer aided analysis of medical images. In particular, the present invention relates to a fast method for detecting lines in medical images. [0002] Line detection is an important first step in many medical image processing algorithms. For example, line detection is an important early step of the algorithm disclosed in U.S. patent application Ser. No. 08/676,660, entitled “Method and Apparatus for Fast Detection of Spiculated Lesions in Digital Mammograms,” filed Jul. 19, 1996, the contents of which are hereby incorporated by reference into the present application. Generally speaking, if the execution time of the line detection step can be shortened, then the execution time of the overall medical image processing algorithm employing that line detection step can be shortened. [0003] In order to clearly illustrate the features and advantages of the preferred embodiments, the present disclosure will describe the line detection algorithms of both the prior art and the preferred embodiments in the context of the computer-assisted diagnosis system of U.S. patent application Ser. No. 08/676,660, supra. Importantly, however, the scope of the preferred embodiments is not so limited, the features and advantages of the preferred embodiments being applicable to a variety of image processing applications. [0004]FIG. 1 shows steps performed by a computer-assisted diagnosis unit similar to that described in U.S. patent application Ser. No. 08/676,660, which is adapted to detect abnormal spiculations or lesions in digital mammograms. At step [0005] At step [0006] Generally speaking, it is to be appreciated that the advantages and features of the preferred embodiments disclosed infra are applicable independent of the size and spatial resolution of the digital mammogram image that is processed. Nevertheless, for clarity of disclosure, and without limiting the scope of the preferred embodiments, the digital mammogram images in the present disclosure, which will be denoted by the symbol I, will be M×N arrays of 12-bit gray scale pixel values, with M and N having exemplary values of 1000 and 1250, respectively. [0007] At step [0008] At step [0009] Finally, at step [0010] One of the desired characteristics of a spiculation-detecting CAD system is high speed to allow processing of more x-ray mammograms in less time. As indicated by the steps of FIG. 1, if the execution time of the line and direction detection step [0011] A first prior art method for generating line and direction images is generally disclosed in Gonzales and Wintz, [0012] The above filter-bank algorithms are computationally intensive, generally requiring a separate convolution operation for each orientation-selective filter in the filter bank. Additionally, the accuracy of the angle estimate depends on the number of filters in the filter bank, and thus there is an implicit tradeoff between the size of the filter bank (and thus total computational cost) and the accuracy of angle estimation. [0013] A second prior art method of generating line and direction images is described in Karssemeijer, “Recognition of Stellate Lesions in Digital Mammograms,” [0014] The Karssemeijer algorithm uses scale space theory to provide an accurate and more efficient method of line detection relative to the filter-bank method. More precisely, at a given level of spatial scale σ, Karssemeijer requires the convolution of only three kernels with the digital mammogram image I, the angle estimation at a pixel (i, j) then being derived as a trigonometric function of the three convolution results at (i, j). [0015]FIG. 2 shows steps for computing line and direction images in accordance with the Karssemeijer algorithm. At step [0016] At step [0017] At step [0018] At step [0019] Each of the line operator functions W [0020] Subsequent steps of the Karssemeijer algorithm are based on a relation shown in Koenderink, supra, which shows that an estimation function W [0021] As indicated by the above definition, the estimation function W θ [0022] Thus, at step [0023] At step [0024] Optionally, in the Karssemeijer algorithm a plurality of spatial scale values σ1, σ2, . . . , σn may be selected at step [0025] Although it is generally more computationally efficient than the filter-bank method, the prior art Karssemeijer algorithm has computational disadvantages. In particular, for a given spatial scale parameter σ, the Karssemeijer algorithm requires three separate convolutions of N [0026] Accordingly, it would be desirable to provide a line detection algorithm for use in a medical imaging system that is less computationally intensive, and therefore faster, than the above prior art algorithms. [0027] It would further be desirable to provide a line detection algorithm for use in a medical imaging system that is capable of operating at multiple spatial scales for detecting lines of varying widths. [0028] It would be even further desirable to provide a line detection algorithm for use in a medical imaging system in which, as the scale of interest grows, the computational intensity grows at a rate less than the rate of growth of the square of the scale of interest. [0029] These and other objects are provided for by a-method and apparatus for detecting lines in a medical imaging system by filtering the digital image with a single-peaked filter, convolving the resultant array with second order difference operators oriented along the horizontal, vertical, and diagonal axes, and computing direction image arrays and line image arrays as direct scalar functions of the results of the second order difference operations. Advantageously, it has been found that line detection based on the use of four line operator functions can actually require fewer computations than line detection based on the use of three line operator functions, if the four line operator functions correspond to the special orientations of 0, 45, 90, and 135 degrees. Stated another way, it has been found that the number of required computations is significantly reduced where the aspect ratio of the second order difference operators corresponds to the angular distribution of the line operator functions. Thus, where the second order difference operators are square kernels, having an aspect ratio of unity, the preferred directions of four line operator functions is at 0, 45, 90, and 135 degrees. [0030] In a preferred embodiment, a spatial scale parameter is selected that corresponds to a desired range of line widths for detection. The digital image is then filtered with a single-peaked filter having a size related to the spatial scale parameter, to produce a filtered image array. The filtered image array is separately convolved with second order difference operators at 0, 45, 90, and 135 degrees. The direction image array and the line image array are then computed at each pixel as scalar functions of the elements of the arrays resulting from these convolutions. Because of the special symmetries involved, the second order difference operators may be 3×3 kernels. Moreover, the number of computations associated with the second order difference operations may be achieved with simple register shifts, additions, and subtractions, yielding an overall line detection process that is significantly less computationally intensive than prior art algorithms. [0031] In another preferred embodiment, the digital image is first convolved with a separable single-peaked filter kernel, such as a Gaussian. Because a separable function may be expressed as the convolution of a first one dimensional kernel and a second one dimensional kernel, the convolution with the separable single-peaked filter kernel is achieved by successive convolutions with a first one dimensional kernel and a second one dimensional kernel, which significantly reduces computation time in generating the filtered image array. The filtered image array is then convolved with three 3×3 second order difference operators, the first such operator comprising the difference between a horizontal second order difference operator and a vertical difference operator, the second such operator comprising the difference between a first diagonal second order difference operator and a second diagonal second order difference operator, and the third such operator being a Laplacian operator. Because of the special symmetries associated with the selection of line operator functions at 0, 45, 90, and 135 degrees, the direction image array and the line image array are then computed at each pixel as even simpler scalar functions of the elements of the arrays resulting from the three convolutions. [0032] Thus, line detection algorithms in accordance with the preferred embodiments are capable of generating line and direction images using significantly fewer computations than prior art algorithms by taking advantage of the separability of Gaussians and other symmetric filter kernels, while also taking advantage of discovered computational simplifications that result from the consideration of four line operator functions oriented in the horizontal, vertical, and diagonal directions. [0033]FIG. 1 shows steps taken by a computer-aided diagnosis (“CAD”) system for detecting spiculations in digital mammograms in accordance with the prior art. [0034]FIG. 2 shows line detection steps taken by the CAD system of FIG. 1. [0035]FIG. 3 shows line detection steps according to a preferred embodiment. [0036]FIG. 4 shows steps for convolution with second order directional derivative operators in accordance with a preferred embodiment. [0037]FIG. 5 shows line detection steps according to another preferred embodiment. [0038]FIG. 3 shows steps of a line detection algorithm in accordance with a preferred embodiment. At step [0039] At step [0040] By single-peaked filter, it is meant that the filter F is a function with a single maximum point or single maximum region. Examples of such a filter include the Gaussian, but may also include other filter kernels such as a Butterworth filter, an inverted triangle or parabola, or a flat “pillbox” function. It has been found, however, that a Gaussian filter is, the most preferable. The size of the single-peaked filter F is dictated by the spatial scale parameter σ. For example, where a Gaussian filter is used, σ is the standard deviation of the Gaussian, and where a flat pillbox function is used, σ corresponds to the radius of the pillbox. In subsequent steps it is assumed that a Gaussian filter is used, although the algorithm may be adapted by one skilled in the art to use other filters. [0041] At step [0042] Advantageously, because the particular directions of 0, 45, 90, and 135 degrees are chosen, these directional derivative operators are permitted to consist of the small 3×3 kernels shown in Eqs. (7a)-(7d):
[0043] The above 3×3 second order directional derivative operators are preferred, as they result in fewer computations than larger second order directional derivative operators while still providing a good estimate of the second order directional derivative when convolved with the filtered image array I [0044] Subsequent steps are based on an estimation function W [0045] It has been found that the extrema of the estimation function W [0046] At step [0047] At step [0048]FIG. 4 illustrates unique computational steps corresponding to the step [0049] Thus, at step [0050] Thus, it is to be appreciated that in the embodiment of FIGS. 3 and 4 a line detection algorithm is executed using four line operator functions W [0051] For illustrative purposes in comparing the algorithm of FIGS. 3 and 4 with the prior art Karssemeijer algorithm of FIG. 2, let us assume that the operations of addition, subtraction, and register-shifting operation take 10 clock cycles each, while the process of multiplication takes 30 clock cycles. Let us further assume that an exemplary digital mammogram of M×N=1000×1250 is used and that N [0052]FIG. 5 shows steps of a line detection algorithm in accordance with another preferred embodiment. It has been found that the algorithm of FIGS. 3 and 4 can be made even more computationally efficient where the single-peaked filter kernel F is selected to be separable. Generally speaking, a separable kernel can be expressed as a convolution of two kernels of lesser dimensions, such as one-dimensional kernels. Thus, the N [0053] Although a variety of single-peaked functions are within the scope of the preferred embodiments, the most optimal function has been found to be the Gaussian function of Eq. (1), supra. For purposes of the embodiment of FIG. 5, and without limiting the scope of the preferred embodiments, the filter kernel notation F will be replaced by the notation G to indicate that a Gaussian filter is being used:
[0054] At step [0055] In accordance with a preferred embodiment, the sigma of the one-dimensional Gaussian kernel G [0056] At step [0057] In accordance with a preferred embodiment, the sigma of the one-dimensional Gaussian kernel G [0058] Even more advantageously, in the situation where N [0059] In addition to the computational savings over the embodiment of FIGS. 3 and 4 due to filter separability, it has also been found that the algorithm of FIGS. 3 and 4 may be made even more efficient by taking advantage of the special symmetry of the spatial derivative operators at 0, 45, 90, and 135 in performing operations corresponding to steps θ [0060] In the above formulas, the array L is defined as follows: [0061] [0062] As known in the art, the array L is the result of the convolution of I [0063] [0064] Finally, the array D in Eqs. (15) and (16) is defined as follows: [0065] [0066] Accordingly, at step [0067] Finally, at step [0068] It is readily apparent that in the preferred embodiment of FIG. 5, steps [0069] The preferred embodiment of FIG. 5 is even less computationally complex than the algorithm of FIG. 3 and [0070] For illustrative purposes in comparing the algorithms, let us again assume the operational parameters assumed previously: that addition, subtraction, and register-shifting operation take 10 clock cycles each; that multiplication takes 30 clock cycles; that M×N=1000×1250; and that N [0071] Optionally, in the preferred embodiment of FIGS. [0072] As another option, which may be used separately or in combination with the above option of using multiple spatial scale values, a plurality of filter kernel sizes N [0073] The preferred embodiments disclosed in FIGS. [0074] [0075] In the general case where the digital mammogram image I is convolved with a single-peaked filter F at step [0076] Importantly, the constant correction factor p does not actually affect the number of computations in the convolutions of Eqs. (6b), (6d), and (22), but rather is incorporated into later parts of the algorithm. In particular, in the algorithm of FIG. 3, the constant correction factor p is incorporated by substituting, for each instance of W [0077] A computational simplification in the implementation of the constant correction factor p is found where the size of the spatial scale parameter 6 corresponds to a relatively large number of pixels, e.g. on the order of 11 pixels or greater. In this situation the constant correction factor p approaches the value of ½, the sampling distance going up by a factor of {square root}2 and the magnitude of the second derivative estimate going up by the square of the sampling distance. In such case, multiplication by the constant correction factor p is achieved by a simple bitwise right register shift. [0078] As disclosed above, a method and system for line detection in medical images according to the preferred embodiments contains several advantages. The preferred embodiments share the homogeneity, isotropy, and other desirable scale-space properties associated with the Karssemeijer method. However, as described above, the preferred embodiments significantly reduce the number of required computations. Indeed, for one of the preferred embodiments, running time increases only linearly with the scale of interest, thus typically requiring an order of magnitude fewer operations in order to produce equivalent results. For applications in which processing time is a constraint, this makes the use of higher resolution images in order to improve line detection accuracy more practical. [0079] While preferred embodiments of the invention have been described, these descriptions are merely illustrative and are not intended to limit the present invention. For example, although the component kernels of the separable single-peaked filter function are described above as one-dimensional kernels, the selection of appropriate two-dimensional kernels as component kernels of the single-peaked filter function can also result in computational efficiencies, where one of the dimensions is smaller than the initial size of the single-peaked filter function. As another example, although the embodiments of the invention described above were in the context of medical imaging systems, those skilled in the art will recognize that the disclosed methods and structures are readily adaptable for broader image processing applications. Examples include the fields of optical sensing, robotics, vehicular guidance and control systems, synthetic vision, or generally any system requiring the generation of line images or direction images from an input image. Referenced by
Classifications
Legal Events
Rotate |