AUTOMATIC HYPERLINKING OF ENGINEERING DRAWING
DOCUMENTS
This patent claims priority from provisional application no.
201611012961Titled “AUTOMATIC HYPERLINKING OF ENGINEERING
DRAWING DOCUMENTS” filed on 13th April 2016.
TECHNICAL FIELD
[0001] This invention relates, generally, to the field of linking of inter –
related documents, and more particularly, to the field of
hyperlinking, automatically, engineering drawing documents based
on linking information identified therein.
BACKGROUND
[0002] All Engineering Projects require AEC (Architecture, Engineering
&Construction) Drawing Documents such as blue prints and/ or plan
documents before the commencement of project work. These
documents are indispensable and integral to the process of
construction or manufacturing. These projects comprise a lot of
smaller tasks and phase – wise activities which in turn, require the
participation of various persons such as engineers, architects,
electricians, plumbers etc. Engineering drawing documents make it
possible for all these persons to execute their respective tasks
3/48
effectively and thus ensure definitive and proper completion of
project(s).
[0003] In recent times, the nature and complexity of these projects has
increased manifolds which in turn, has impacted the corresponding
engineering drawing documents. Now instead of having a single (or
a countable few) engineering drawing documents, each project
comprises numerous such documents. In certain cases these
multiple documents are further divided and sub-divided so as to
manage each section of a project efficiently.
[0004] Navigation through such large number of drawing sheets and its
sub-sections has therefore, become a tedious and cumbersome
task for itsusers. Accordingly, various systems have been designed
and developed to cater to the needs of the users in this space.
[0005] One class of the state-of-the-art systems in this regard, facilitated
navigation through the engineering drawing sheets by manually
indexing the documents. However, these systems suffered from
vulnerability to human errors whichcould renderthe whole indexing
process futile.
[0006] One other class of the state-of-the-art systems facilitated navigation
through theengineering drawing sheets by reading the linking
information present in the callout regions by using traditional optical
reading techniques. However, these systems often yielded
4/48
inaccurate results due to various technical limitations on the Optical
Character Recognition(OCR) tools, particularly in case of those
engineering drawing sheets that contained high noise and large
graphical content.
[0007] Therefore, it is desirable to propose a solution that realizes and
resolves the disadvantages of the prior art systems, and
achieveseasy navigation through engineering drawing sheets
without being restrained by factors such as high noise in the
engineering drawings, unnecessary marks and graphic components
in the documents, difficulties in extraction of linking information etc.,
and operates at a relatively higher accuracy (about 94 – 95%).
SUMMARY OF INVENTION
[0008] The present invention is intended to address atleast the above
mention problems and/or disadvantages and to provide a suitable
solution. Accordingly, an aspect of the present invention is to
provide a system and a method for automatically hyperlinking
engineering drawing documents for facilitating easy navigability and
referencing.
[0009] In accordance with an aspect of the present invention , a system
tohyperlink, automatically, one or more engineering drawing
documents is provided, the system comprising a pre processing
module configured to accept the one or more engineering drawing
5/48
documents and generate corresponding one or more workable
input documents, a callout identification module, logically
connectible with the pre – processing module, configured to locate
one or more callout regions from within the one or more workable
input documents,a text segregation module, logically connectible
with the callout identification module, configured to separate out
one or more text data from one or more non – text data (i.e. graphic
content) within the one or more callout regions, an alphanumeric
code segregation module, logically connectible with the text
segregation module, configured to read one or more alphanumeric
codes from within the one or more text data, wherein the
alphanumeric codes includes, without limitation, sheet number and/
or indexation information, and a hyperlinking module, logically
connectible with the alphanumeric code segregation module,
configured to establish one or more hyperlinks between one or
more engineering drawing documents.
[0010] In accordance with one other aspect of the present invention, a
method for hyperlinking, automatically, one or more engineering
drawing documents is provided, the method comprising the steps
ofaccepting the one or more engineering drawing documents and
generating corresponding one or more workable input documents,
locating one or more callout regions from within the one or more
workable input documents, separating out one or more text data
6/48
from one or more non – text data within the one or more callout
regions,reading one or more alphanumeric codes from within the
one or more text data, wherein the alphanumeric codes includes,
without limitation, sheet number and/ or indexation information, and
establishing one or more hyperlinks between the one or more
engineering drawing documents.
[0011] Other aspects, advantages and salient features of the invention will
become apparent to those skilled in the art from the following
detailed description, which, taken in conjunction with the annexed
drawings, discloses exemplary embodiments of the invention.
BRIEF DESCRIPTION OF DRAWINGS
[0012] The above and other aspects, features and advantages of certain
exemplary embodiments of the present invention will be more
apparent from the following detailed description taken in
conjunction with the accompanying drawings, in which:
[0013] Fig. 1 illustrates a high level block diagram of a preferred
embodiment of the claimed system,
[0014] Fig. 2 describes a preferred embodiment of the Pre processing
module,
[0015] Fig. 3 illustrates a sample engineering document with callouts,
[0016] Fig. 4 provides for an enlarged view of a sample callout region,
7/48
[0017] Fig. 5 illustrates a preferred embodiment of the callout identification
module,
[0018] Fig. 6 shows a sample representation of an edge image
corresponding to an engineering drawing document,
[0019] Fig. 7 is a table indicating the accuracy with which true circles are
detected relative to centre threshold values,
[0020] Fig. 8 shows some of the problematic callout circular regions which
may be found in engineering drawing documents,
[0021] Fig. 9 is an exemplary representation of a connected graph which is
established between the related workable input documents,
[0022] Fig. 10 illustrates a high – level flow chart of a non – limiting,
exemplary embodiment of a method for hyperlinking one or more
engineering drawing documents,
[0023] Fig. 11 provides for a flow chart illustrating the step of accepting the
one or more engineering drawing documents and generating their
corresponding workable input documents,
[0024] Fig. 12 provides for a specific embodiment of the step of accepting
the one or more engineering drawing documents,
[0025] Fig. 13 describes a flow chart illustrating the step of locating one or
more callout regions from within the workable input documents,
[0026] Fig. 14 illustrates an exemplary flow chart of the step of establishing
hyperlinks between the one or more engineering drawing
documents,
8/48
[0027] Fig.15 illustrates an exemplary computer system in which various
embodiments of the present invention may be implemented; and
[0028] Fig. 16 illustrates an exemplary representation of the claimed
invention comprising the computer system in accordance with Fig.
15, and a scanner hardware.
DETAILED DESCRIPTION
[0029] The following description with reference to the accompanying
drawings is provided to assist in a comprehensive understanding of
exemplary embodiments of the invention as defined by the claims
and their equivalents. Those skilled in the art would appreciate that
the following description is for illustration purpose only and not for
the purpose of limiting the invention as defined by the appended
claims and their equivalents. Those of ordinary skill in the art will
also understand that various changes and modifications of the
embodiments described herein may be made without departing
from the scope and spirit of the invention.
[0030] Additionally, the terms and words used in the following description
and claims are no limited to the bibliographical meanings, but are
merely used by the inventor to enable a clear and consistent
understanding of the invention. It is to be understood that the
similar forms “a”, “an”, and “the” include plural referents unless the
context clearly dictates otherwise.
9/48
[0031] Fig. 1 to Fig. 16, discussed below, are non-limiting illustrations
which may be used to explain and describe the invention used to
explain and describe the invention disclosed herein. Persons skilled
in the art will appreciate that the purpose of these figures is to
provide clarity on the concepts associated with the various technical
embodiments of this invention.These figures may include block
diagrams and/or flowcharts and/or other simplistic representations
and may not always indicate the real time operations taking place in
the invention.
[0032] Fig. 1 illustrates a high level block diagram of a preferred
embodiment of the claimed system[100]. As indicated in the
diagram this embodiment, largely comprises a Pre processing
Module [110],which in the broadest sense, accepts one or more
engineering drawing documents and generates their corresponding
workable input documents, a Callout Identification Module
[120],which in the broadest sense, locates callout regions from the
the workable input documents, a Text Segregation Module
[130],which in the broadest sense, separates out text data from the
non – text data (i.e. graphical content), within callout regions, an
Alphanumeric Code Segregation Module [140], which in the
broadest sense, reads one or more alphanumeric codes ( i.e.
linking information) from within the text data, and a hyperlinking
Module [150], which in the broadest sense, is configured to
10/48
establish one or more hyperlinks within the engineering drawing
document based on the alphanumeric codes; the modules being
logically connectible with each other.
[0033] In accordance with Fig. 1, the one or more engineering drawing
documents are first inputted into a pre processing module [110]
which generates one or more workable input documents in systemingestible
formatswith their images logically separated and the
images categorized into distinct images segments, the segments
bearing homogeneity in terms of properties including but not limited
to greyvalue or textures. These workable input documents are then
moved into a callout identification module [120] wherein callout
regions are identified in the engineering drawing documents by
serially pursuing hole detection, circle detection and callout
detection mechanisms. The callout regions thus identified, are then
moved into a text segregation module [130] and alphanumeric code
segregation module [140] respectively, wherein text data is
separated out from the callout regions, and, alphanumeric linking
information such as sheet numberand/ or indexation information is
read from the text data. The alphanumeric linking information is
then inputted into a hyperlinking module [150] that establishes
hyperlinks between the one or more engineering drawing
documents.
11/48
[0034] Fig. 2 describes a preferred embodiment of the Preprocessingmodule
within the claimed system [200]. As already
stated this module [110, 200] is focussed at accepting the one or
more engineering drawing documents and generating their
corresponding workable input documents. This conversion of the
engineering drawing documents involves a three – part transition,
viz. (a) transforming the engineering drawing documents into one or
more system – ingestible formats, (b) identifying, distinctly, the
images from the system - ingestible formats, and (c) partitioning the
one or more images into one or more image segments. Each of
these transitionsoccurs in a serial fashion, by individually invoking
the units within the pre – processing module [200], i.e. Conversion
Unit [210], Image Creation Unit[220] and Segmentation Unit [230],
respectively.
[0035] The Conversion Unit [210] helps achieve the first transition of
converting the engineering drawing documents into system -
ingestible formats. These system - ingestible formats, for the
purpose of this disclosure, mean such formats which are easily
readable and comprehendible by the claimed system.These
formats, without limitation, may include grey scale formats, RGB
formats, binary formats and such other like formats.
[0036] As per an exemplary embodiment, the conversion unit [210]
transforms the engineering drawing documents into corresponding
12/48
binary format documents. This transformation is achieved by
converting the engineering drawing documents into their
corresponding RGB format documents, which in turn are converted
into their equivalent greyscale format documents. The greyscale
format documents are then transformed into their binary format
documents.
[0037] The Image Creation Unit [220]helps achieve the second transition
of identifying, distinctly, the one or more images from within the
system – ingestible formats. In a preferred embodiment, the image
creation unit [220] receives the one or more binary format
documents, from the conversion unit[210] and, divides and
distinctlyidentifies, the one or more constituent images therein.
[0038] The Segmentation unit [230] helps achieve the third transition of
partitioning the images identified in the system – ingestible formats
in one or more image segments. Image segments, for the purpose
of disclosure, mean those disjoint regions within images which
possess homogeneity in terms properties including but not limited
to grey value or texture. This partitioning into one or more image
segments, is done to simplify and/or change the representation of
an image into something more meaningful and easier to analyse,
e.g. separating out the images into their distinct objects and
corresponding boundaries (lines, curves etc.).
13/48
[0039] In a preferred embodiment, the segmentation unit [230] receives
the one binary format documentswith their images distinctly
identified, from the image creation unit [220], and segments them
into their corresponding objects andboundaries (lines, curves, etc.)
by applying Otsu's technique whichassigns differentialpixels to
foreground and background of the one or more binary format
documents, based on varying greyscale intensity. In other
embodiments, segmentation of images may be achieved by other
named and unnamed techniques, including thewatershed technique
and its variations. Once workable input documents are generated
from the pre processing module, they are subjected to the next task
of locating callout regions.
[0040] Fig. 3and Fig. 4 substantiate upon callout regions. Fig. 3 illustrates
a sample engineering document with callout regions in it, and Fig. 4
provides for an enlarged view of a sample callout region. As may be
seen in these figures, callout regions are usually circular or near
circular symbols which are present at different locations within the
engineering drawing documents. In an exemplary representation,
these callout regions may contain a horizontal and a vertical line
that approximately divides the circles into two halves, each half
bearing alphanumeric characters which, inter alia represent
‘Section information’that indicates the location of a specific
segment, and/ or ‘Destination Sheet strings’that indicates the name
14/48
and number of that sheet which is to be referred by that specific
segment.
[0041] Fig. 5 illustrates a preferred embodiment of the callout identification
module[500], configured to locate one or more callout regions from
within the workable input documents. This location of callout
regions is achieved by systematic application of image processing
techniques at three levels (a) identification of holes within workable
input documents, (b) identification of circular and/ or near circular
symbols from within the holes and, (c) identification of callouts from
within the circular and/ or near circular symbols. The three stages of
callout region identification are performed serially by systematic
invocation of the units within the callout identification module [500],
i.e. hole detection unit [510], circle detection unit [520] and callout
detection unit [530], respectively.
[0042] The hole detection unit [510] is responsible for the identification of
holes from within workable input documents.However, to
accomplish the identification of holes, the hole detection unit [510]
first refines the workable input documents by transforming the
images within the workable input documents into corresponding
boundary components, thus reducing the number of foreground
pixels in a graphics rich image.
[0043] In a preferred embodiment of refinement of the workable input
documents, the hole detection unit [510] receives the binary format
15/48
documents, smoothens the one or more images from within the
binary format documents, by applying one or more filters including
without limitation, one or more Gaussian filters, obtains one or more
edge gradients and one or more pixel directions corresponding to
the one or more smoothened images, creates the one or more
edges based on the one or more edge gradients and the one or
more pixel directions, and connects the edges to obtain one or
more preliminary edge images. Unwanted pixels that do not
constitute the one or more edges in the preliminary edge images,
and further those edges within the preliminary edge images which
do not possess proper edge gradient intensity and/ or which are
unconnected, are then removed to obtain final edge images. Fig. 6
shows a sample representation of an edge image corresponding to
an engineering drawing document [600].
[0044] After the conversion from binary image to edge image, the Hole
Detection Unit [510] conducts a connected component (CC)
analysis, post which all graphic components which do not suit one
or more criteria of the circular or near circular shapes of callout
regions are rejected. For instance, in one case, all graphically
connected components having height to width ratio more than a
specified threshold may be discarded for further processing and
scrutiny. In one other case, all graphically connected components
having density (i.e. black to white pixel ratio) lower or higher than a
16/48
specified threshold may get rejected. Summarily, the end result is
that all unnecessary and unwanted graphics components are
discarded and only those graphical components, bearing holes, are
retained for further processing.
[0045] The circle detection unit [520] is responsible for the identification of
circular and/ or near circular symbols from within identified holes. In
a preferred embodiment, the Circle Detection Unit [520]receives the
holes detected from the hole detection unit and subjects the holes
to a “Modified Hough Transform” with randomized point selection
for identifying and selecting circles.A description of the “Modified
Hough Transform” used in the instant context for detection of
Circles is explained hereinunder.
[0046] Circle may be represented in two dimensional space as ,
(x – a)2 + (y – b)2 = r2.......(1),
where, a, b represent the center of the circle and r is the radius.
[0047] Most of the architects follow the same standards for making callouts
and therefore the callout circles are almost similar in size. Due to
similarity in size, the radius range for finding the circles is also fixed.
Accordingly, for a fixed radius r, the parameter space is reduced to
two dimension (position of circle center) i.e. for each point (x, y) on
the original circle, all the circles with center at (x, y) and radius r,
may be found according to (1), and the intersection of all these
17/48
circles in parameter space would be the center (a, b) of the original
circle.
[0048] In light of supra explanation, an accumulator matrix with initial
values of all elements set to zero, may be configured for tracking
intersections of all circles such that for for each point a circle
passes through, the matrix increases the vote value. In such
arrangement, wherever the vote value reaches its highest, the
matrixretrieves that point and calculates its local maxima to
determine the center of a circle, which is then used to identify a
circle with a fixed radius “r” around the determined centerof the
circle.
[0049] However, as the callout shapes are not perfect circles, more than
one circle may be detected at a single circle center. Accordingly,
these circles may often encompass improper or false circles not
bearing linking information, which may be eliminated during further
processing. These improper or false circles, may however be
minimized by maintaining the voting threshold at an optimum value,
thereby attaining maximum accuracy. Person skilled in the art
would appreciate that other than voting threshold for centers of
circles, various other thresholds/ parameters such as Minimum
Radius, Maximum Radius, and Minimum distance between two
centers, may also be used for implementing this technique.
18/48
[0050] Fig. 7 is a table representing the accuracy with which true
circles(i.e. circles which are not false or improper) are detected
relative to center threshold values. From the figure it may be
noticed that as the center threshold value is decreased, more and
more circles are detected and the accuracy increases, however as
the center threshold value is further decreased, more and more
false circles are detected, which are then required to be detected
and removed in the subsequent stages. This clearly shows an
inverse proportionality between the number of detected circles “N”
and the center threshold value “CT”. The relationship may be
depicted as below -
CT 1/N…….(2)
[0051] The callout detection unit [530] is responsible for the identification
callout regions from amongst the circles identified in the circle
detection unit [520]. More particularly, this module eliminates the
false/ imperfect circles that may have passed through the circle
detection unit [520], and obtains the final set of callouts.
[0052] These false circles may be false positives i.e. circles which are
circular in shape but are not same as the callout shape, or false
negatives i.e. circular shapes which are not circular at all. In a
preferred embodiment, the callout detection unit, based on the
reasonable presumption that callout regions cannot intersect or
overlap, removes the false/ imperfect circles by taking the minimum
19/48
distance threshold between two local maxima of a circle to be the
diameter of the circle,and in light of the equation (2) state above,
calculatively increases the voting values/ center thresholds in order
to decrease the number of false circles and retain only the true
circles.
[0053] Text Segregation Module [130], is the next module which is serially
and logically, connectible with the callout identification module
[120], and which in the broadest sense, separates out the text data
from the non - text data (i.e. graphical content) within the callout
regions. The Text Segregation Module [130] achieves this by
subjecting the input callout regions to one or more edits and
modifications as per image processing techniques and
mathematical modifications. This separation of text from graphical
components is required so that the OCR may easily and effectively
recognize the alphanumeric code which constitutes the linking
information.
[0054] Fig. 8shows some of the problematic callout circular regions which
may be found in engineering drawing documents. The Text
Segregation Module [130]works on each of these callout regions,
subjectively and intuitively, and extracts the text data on a per case
basis.
[0055] For instance, in case of disoriented callout regions, the text
segregation module[130] may obtain the callout’s orientation by
20/48
detecting the straight line through the centre of the circle (either
horizontal or vertical). In this case, the text segregation module
[130], may directly scan through the circles having horizontal lines,
and rotate the circles by 90 degrees in case of circles with vertical
lines. Once the straight lines are detected, all the linear
components including horizontal and vertical lines may be removed
from the image, to retain only the text data. Person skilled in the art
would appreciate that the threshold for removal of the linear
components is based on the height and width of the image, in order
to prevent the accidental removal of characters with large linear
components.
[0056] In accordance with Fig. 8, some problematic callout circular regions
may have their text data (e.g. Destination sheet characters)
touching the circle boundary. In such cases, a connected
component analysis (CCA) may be performed on the components
of image and the largest component i.e. the callout circular region
itself may be removed to obtain optimum text data. More
specifically, in such cases, the callout regions may be converted
into a thin image, thereby preserving the connectivity of the
components. The object pixels may then be traversed in downward
direction. As soon as a direction change in the upward direction is
encountered, the pixel may be cut out from that point and
downward traversal be continued. After reaching half the width of
21/48
the image, the same process may be started from the right – top of
the image, by continuing to move in the downward direction. After
all the joints between the characters and the circle boundary are
disconnected, the circle may be removed without letting the
characters from getting removed due to the connectivity.
[0057] In one other embodiment, the text segregation module [130] may
further be enabled to remove noise components from within the
present in the images. In such cases, a horizontal line
segmentation using an object pixel histogram may be conducted as
regards thegraphical components which are smaller in height than
the text data written within the circle. A region of interest may be
selected as the region having the histogram peak and all the
objects which have a histogram value larger than 80% of the
histogram peak and which are in the same line as that of the peak
are selected. In other variants of this embodiment, Height – width
ratio and component density based approaches may be used to
remove smaller components and components with very high black
pixel density.
[0058] In other embodiment and variants, the text segregation module
[130] may discard the callout circular regions based on non –
satisfaction of one or morecallout requirements or threshold values.
Examples may include discarding of callouts where there is no
22/48
Destination sheet information, or discarding of callouts whose
density is less than a minimum density threshold value.
[0059] Alphanumeric Code Segregation Module [140] is the next module in
the claimed system [100] which is serially and logically connectible
to the Text Segregation Module. This module in the broadest sense
is configured to read alphanumeric codes from the text data
extracted by the Text Segregation Module [130]. In context of this
disclosure, alphanumeric codes may be any information which
could be used for establishing hyperlinks between engineering
drawing documents. These alphanumeric codes may include
without limitation, sheet number, section number, and indexation
number.
[0060] In a preferred embodiment the alphanumeric code segregation
module [140] uses a customized Optical Character Recognition
(OCR) engine having a support vector machine classifier with
capability of non – linear classification may be used, wherein the
classification may be based on the proper selection of features
which could either be local or global in nature. Local features may
involve windowing the image whereas global features may include
characteristics of the whole image. In accordance with this
customized OCR, the image of the character may be divided into 5
x 5 block local features along with 4 border pixels in 4 directions,
thereby obtaining 5x5x4=100 features which may be used to
23/48
identify and extract the alphanumeric characters. In certain cases,
global features like aspect ratio, longest vertical run, normalized
height of the leftmost, rightmost and lowermost black pixel, Euler
number etc. may also be used for more accurate extraction of the
alphanumeric characters.
[0061] The Hyperlinking Module [150] is the next and the last module,
which is serially and logically connectible, with the alphanumeric
code segregation module [140] that establishes hyperlinks between
the engineering drawing documents. Once the alphanumeric codes
are received by the hyperlinking module, it firstly, identifies all local
files (i.e. engineering sheets within the same project) related with
each other based on the codes. For instance, if the alphanumeric
code which reaches the hyperlinking module is the sheet number of
the destination sheet i.e. the number of the sheet to which the
current sheet refers, then this module would reiteratively run
through the sheet numbers of all destination sheets, and as a
consequence identify all related sheets which may have been read
through, pursuant to the reiteration.
[0062] Fig. 9 is an exemplary representation of a connected graph which
may be established between one or more related workable input
documents. In accordance with the figure, the bold dots represent
the engineering sheets, the alphanumeric text near the dots
represents the sheet number, the arrows represent a referential
24/48
relationship (i.e. indexation) between the engineering sheets, and
the direction of arrows represents order of indexing i.e. number
from which the arrow originates is that sheet number which has to
be hyperlinked to such other sheet number bearing the number at
which the arrow ends.
[0063] Once the connected graphs are created, the foundation for easy
access and navigability between the related workable input
documents is laid down. In accordance with Fig.9,after the
engineering drawing documents are indexed according to their
sheet numbers, then when a particular document AS-101 bears a
code AS-103 in its callout region, then a connection is automatically
established between drawings AS-101 and drawing AS-103.
[0064] Once connection graphs between these drawing sheets are found,
the linking information is simply burnt into the original document,
and clickable links are formed which enable navigation among the
various plan documents and engineering drawing documents in a
project
[0065] Fig. 10 illustrates a high – level flow chart of a non – limiting,
exemplary embodiment of a method for hyperlinking one or more
engineering drawing documents [1000]. As may be seen in the
figure, this preferred embodiment of the method [1000] largely
comprises the steps of (a) accepting the one or more engineering
drawing documents and generating their corresponding workable
25/48
input documents [1010], (b) locating callout regions from the
workable input documents [1020], (c) separating out text data from
non – text data (i.e. graphical content), with the callout regions
[1030], (d) reading alphanumeric codes (i.e. linking information)
from within the text data [1040], and (e) establishing hyperlinks
between one or more engineering drawing documents based on the
alphanumeric codes[1050].
[0066] Fig. 11 provides for a flow chart illustrating the step of accepting the
one or more engineering drawing documents and generating their
corresponding workable input documents [1100], the step being
characterized by (a) transforming the engineering drawing
documents into one or more system – ingestible formats, wherein
the system - ingestible formats mean such formats which are easily
readable and comprehendible by the claimed method [1110], (b)
identifying, distinctly, the images from the system - ingestible
formats [1120], and (c) partitioning the one or more images into one
or more image segments [1130].
[0067] Fig. 12 provides for an exemplary embodiment of the step of
accepting the one or more engineering drawing documents [1200],
as disclosed in Fig. 11. In accordance with this figure, the step of
accepting the one or more engineering drawing documents and
generating their corresponding workable input documents[1200]
involves the sub – step of transforming the engineering drawing
26/48
documents into their corresponding binary format documents [1210]
by firstly converting the engineering drawing documents into their
corresponding RGB format documents [1203], then converting the
RGB format documents into their equivalent greyscale format
documents [1206], and finally converting the greyscale format
documents into their corresponding binary formats [1209].
Thereafter, the next sub – step of distinctly, identifying, the one or
more constituent images within the binary format documents[1220].
Lastly, the sub – step of partitioning the images identified in
thebinary format documents into one or more image segments is
performed [1230]. This partitioning into image segments may be
achieved by applying Otsu's technique which assigns pixels to
foreground and background of the one or more binary format
documents, based on varying greyscale intensity, watershed
technique and/ or such other named and unnamed techniques.
[0068] Once the one or more workable input documents are generated,
they are subjected to the step of locating callout regions [1120].
These callout regions are usually circular or near circular symbols
which are present at different locations within the engineering
drawing documents. In an exemplary representation, these callout
regions may contain a horizontal and a vertical line that
approximately divides the circles into two halves, each half bearing
alphanumeric characters which, inter alia represent ‘Section’
27/48
information’ which indicates the location of a specific segment and
‘Destination Sheet strings’ which indicates the name and number of
that sheet which is to be referred by that specific segment.
[0069] Fig. 13describes a flow chart illustrating the step of locating one or
more callout regions from within the workable input
documents[1300]. The locating [1300] being characterized by the
sub – steps of (a) identifying holes within the workable input
documents[1310], (b) identifying circles from within the holes [1320]
and, (c) identifying callouts from within the circles [1330]. Each of
these sub - steps are systematically, performed and executed, one
after another. In accordance with this figure, the first sub – step of
identifying holes within the workable input documents [1310]
comprises, receiving binary formatted documents corresponding to
engineering drawing documents [1301], smoothening the images
within the binary formatted document by applying filters such as
Gaussian Filter [1302], obtaining an edge gradient and a pixel
direction for the smoothened images and forms edges based on the
edge gradient and pixel direction [1303], connecting the edges to
obtain edge images [1304], and conducting a connected
component (CC) analysis [1305], post which all graphic
components which do not suit one or more criteria such as height to
width ratio and black to white pixel ratio, are rejected. Once the
holes are detected in the workable input documents, they are
28/48
subjected to the second sub - step of identifying circles from within
the holes [1320], which inter alia, is achieved by implementing a
“Modified Hough Transform” technique, details of which have
already been provided. Lastly the sub – step of identifying callout
regions is executed, wherein all false/ imperfect circles that may
have erroneously passed through the previous sub – step of
identifying circles from within the holes are discarded, and only
those circles which are callout regions i.e. which bear linking
information are retained.
[0070] Once the callout regions are identified from within the workable
input documents, the step of separating out the text data from
within the callouts is triggered [1130]. This step [1130] is executed
by subjecting the input circles to one or more edits and
modifications as per image processing techniques and
mathematical modifications.
[0071] Once the text data is received from within the one or more callout
regions, the step of reading the one or more alphanumeric codes
from the text data received thereto is triggered [1140]. This reading
of one or more alphanumeric codes [1140] is pivotal as the codes
yield the linking information required for hyperlinking the
engineering drawing documents. In context of this disclosure,
alphanumeric codes may be any information which could be used
for establishing hyperlinks between engineering drawing
29/48
documents, including without limitation, sheet numbers, section
numbers, and indexation numbers. In a preferred embodiment of
this step of reading the one or more alphanumeric codes [1140],a
customized optical character recognition (OCR) engine using
support vector machine classifier with capability of non – linear
classification may be used, wherein the classification may be based
on the proper selection of features which could either be local or
global in nature. Local features may involve windowing the image
whereas global features may include characteristics of the whole
image such as aspect ratio, longest vertical run, normalized height
of the leftmost, rightmost and lowermost black pixel, and Euler
number.
[0072] Once the one or more alphanumeric codes i.e. linking information is
realized from the callout regions, the last step establishing
hyperlinks between the one or more engineering drawing
documents is triggered [1150].
[0073] Fig. 14 illustrates an exemplary flow chart of the step of establishing
hyperlinks between the one or more engineering drawing
documents, [1400]. In accordance with this step [1400], firstly, all
related engineering drawing documents within a given engineering
project are identified [1410]. To illustrate, suppose the
alphanumeric code realized from the callout regions is the sheet
number of the destination sheet i.e. the number of the sheet to
30/48
which the current sheet refers, then this sub – step would
reiteratively run through the sheet numbers of all destination
sheets, and continue the reiteration until all related sheets are
identified.Once the related engineering drawing documents are
identified, the next sub – step within the step of establishing
hyperlinks is triggered, which is creating connected graphs within
the related engineering drawing documents [1420]. Once the
connected graphs are created, the foundation for easy access and
navigability between the related workable input documents is laid
down. Post - creation of the connected graphs, the last – step is
executed, the sub – step comprising burning the linking information
into the original document, and forming clickable links [1430] which
enable navigation between the various engineering drawing
documents within a project.
[0074] Fig.15illustrates an exemplary computer system [1500] in which
various embodiments of the present invention may be implemented.
The computer system [1500] comprising a processor [1504] and a
memory [1506]. The processor [1504] executes program
instructions and may be a real processor. The processor [1504]
may also be a virtual processor. The computer system [1500] is not
intended to suggest any limitation as to scope of use or functionality
of described embodiments. For example, the computer system
[1500] may include, but not limited to, a general-purpose computer,
31/48
a programmed microprocessor, a micro-controller, a peripheral
integrated circuit unit, and other devices or arrangements of
devices that are capable of implementing the steps that constitute
the method of the present invention. In an embodiment of the
present invention, the memory [1506] may store software for
implementing various embodiments of the present invention. The
computer system [1500] may have additional components. For
example, the computer system [1500] includes one or more
communication channels [1508], one or more input devices [1510],
one or more output devices [1512], and storage [1514]. An
interconnection element (not shown) such as a bus, controller, or
network, interconnects the components of the computer system
[1500]. In various embodiments of the present invention, operating
system software (not shown) provides an operating environment for
various software executing in the computer system [1500], and
manages different functionalities of the components of the
computer system [1500].
[0075] The communication channel(s) [1508] allow communication over a
communication medium to various other computing entities. The
communication medium provides data such as program
instructions, or other data in a communication media. The
communication media includes, but not limited to, wired or wireless
32/48
methodologies implemented with an electrical, optical, RF, infrared,
acoustic, microwave, blue tooth or other transmission media.
[0076] The input device(s) [1510] may include, but not limited to, a
keyboard, mouse, pen, joystick, trackball, a voice device or any
another device that is capable of providing input to the computer
system [1500]. In an embodiment of the present invention, the input
device(s) [1510] may be a sound card or similar device that accepts
audio input in analog or digital form. The output device(s) [1512]
may include, but not limited to, a user interface on CRT or LCD,
printer, speaker, CD/DVD writer, or any other device that provides
output from the computer system [1500].
[0077] The storage [1514] may include, but not limited to, magnetic disks,
magnetic tapes, CD-ROMs, CD-RWs, DVDs, flash drives or any
other medium which may be used to store data and may be
accessed by the computer system [1500]. In various embodiments
of the present invention, the storage [1514] contains program
instructions for implementing the described embodiments.
[0078] The present invention may suitably be embodied in the computer
system [1500] or any other similar device, whereinthe claimed
method and system may typically be implemented as a set of
program instructions. The set of program instructions may be a
series of computer readable codes stored on a tangible medium,
such as a computer readable storage medium storage [1514], for
33/48
example, diskette, CD-ROM, ROM, flash drives or hard disk, or
transmittable to the computer system [1500], via a modem or other
interface device, over either a tangible medium, including but not
limited to optical or analogue communications channel(s) [1508].
The implementation of the invention may be in an intangible form
using wireless techniques, including but not limited to microwave,
infrared, blue tooth or other transmission techniques. These
instructions may be preloaded into a system or recorded on a
storage medium such as a CD-ROM, or made available for
downloading over a network such as the internet or a mobile
telephone network. The series of computer readable instructions
may embody all or part of the functionality previously described
herein.
[0079] Fig. 16illustrates an exemplary representation of the claimed
invention [1600], in accordance with which one or more hard copies
of engineering drawing documents [1610] are put into hardware
devices, including without limitation, scanners and photographic
devices [1620], so as to generate image files of the engineering
drawing documents. These image files of scanned engineering
drawing documents are then inputted into the computer system
[1630], that executes the invention disclosed herein, particularly the
claimed system and method. Person skilled in the art would
appreciate that in other embodiments of the present invention,
34/48
scanned copy of the document may be made available to the
claimed system and method [1640], by internal storage or cache of
thecomputer system or via communication channels of the
computer system itself.
[0080] The present invention may be implemented in numerous ways
including as a system and a method which may be executed by a
computer system as described above or as a system and a method
wherein programming instructions are communicated from a remote
location, via a communication network.
[0081] In relation to the preceding specification, it is reiterated that the
present disclosure and its advantages have been described with
reference to exemplary embodiments and that, a person of ordinary
skill in the art would appreciate that various modifications and
changes may be made, without departing from the scope of the
present disclosure, as set forth in the appended claims and their
equivalents. Furthermore, it is re-emphasized that the preceding
specification and figures are to be regarded as illustrative examples
of the present disclosure, rather than in restrictive sense. All such
possible modifications are intended to be included within the scope
of the present invention.
35/48
I/ We Claim:
1. A system to hyperlink one or more engineering drawing documents,
the system comprising -
a pre processing module configured to accept the one or
more engineering drawing documents and generate
corresponding one or more workable input documents,
wherein the pre processing module is characterized by -
a conversion unit that transforms the one or more
engineering drawing documents into one or more
system – ingestible formats,
an image creation unit that distinctly, identifies, the
one or more system – ingestible engineering drawing
documents into one or more images, and
a segmentation unit that partitions the one or more
images into one or more image segments, wherein
the image segments are disjoint regions bearing
homogeneity in terms of properties including without
limitation, grey value or texture,
further wherein, the transformation of the one or more
engineering drawing documents into one or more system –
ingestible formats, the division of the one or more system –
36/48
ingestible engineering drawing documents into one or more
images, and the partitioning of the one or more images into
one or more image segments, yields the workable input
documents.
a callout identification module, logically connectible with the
pre processing module, configured to locate one or more
callout regions from within the workable input documents,
wherein the callout identification module is characterized by -
a hole detection unit that identifies one or more holes
in the one or more workable input documents,
a circle detection unit that identifies one or more
circular symbols and/ or near circular symbols from
within the one or more holes, and
a callout detection unit that identifies one or more
callout regions from within the one or more circular
symbols and/ or near circular symbols,
a text segregation module, logically connectible with the
callout identification module, configured to separate out one
or more text data fromwithin the one or more callout regions,
an alphanumeric code segregation module, logically
connectible with the text segregation module, configured to
37/48
read one or more alphanumeric codes from within the one or
more text data, wherein the alphanumeric codes includes
without limitation, sheet number and/ or indexation
information; and
a hyperlinking module, logically connectible with the
alphanumeric code segregation module, configured to
establish one or more hyperlinks between the one or more
engineering drawing documents.
2. The system as claimed in claim 1, wherein the pre processing
module comprises the conversion unit that transforms the one or
more engineering drawing documents into one or more binary
format documents by -
converting the one or more engineering drawing documents into
corresponding one or more RGB format documents,
converting the one or more RGB format documents into
corresponding one or more greyscale format documents; and
converting the one or more greyscale format documents into
corresponding one or more binary format documents.
38/48
3. The system as claimed in claim 1, wherein the pre processing
module comprises the segmentation unit that partitions the one or
more binary format documents into the one or more image
segmentsby assigning differential pixels to the background and
foreground of the one or more binary format documents, based on
varying greyscale intensity.
4. The system as claimed in claim 1, wherein the pre processing
module comprises the segmentation unit that partitions the one or
more binary format documents into the one or more image
segments by techniques, including without limitation, watershed
technique and/or its variations.
5. The system as claimed in claim 1, wherein the callout identification
module comprises the hole detection unit that identifies the one or
more holes from within the one or more binary format documents by
-
smoothening the one or more images from within the binary
format documents, by applying one or more filters including
without limitation, one or more Gaussian filters,
39/48
obtaining one or more edge gradients and one or more pixel
directions corresponding to the one or more smoothened
images,
creating one or more edges based on the one or more edge
gradients and the one or more pixel directions,
connecting the one or more edges to obtain one or more
preliminary edge images,
removing any unwanted pixels that do not constitute the one or
more edges corresponding to the one or more preliminary edge
images,
determining one or more final edge images from the preliminary
edge images by eliminating those edges which possess
improper edge gradient intensity and/ or which are unconnected;
and
identifying one or more hole bearing edge images from the one
or more final edge images by eliminating those one or more final
edge images that do not match one or more criteria including
without limitation, height to width ratio and/ or black to white
pixel ratio.
40/48
6. The system as claimed in claim 1, wherein the callout identification
module comprises the circle detection unit that identifies the one or
more circular symbols and/ or near circular symbols from within the
one or more holes, by an accumulator matrix that uses reference
parameters including without limitation, voting threshold for centre
of circle, minimum radius, maximum radius, and minimum distance
between two centers of circles.
7. The system as claimed in claim 1, wherein the callout identification
module comprises the callout detection unit that identifies one or
more callout regions from within the one or more circular symbols
and/ or near circular symbols by –
pre setting the one or more reference parameters of the circular
and/ or near circular symbols to an optimum value, the reference
parameters including voting threshold for centre of circle,
minimum radius, maximum radius, and minimum distance
between two centers of circles; and
discarding the one or more circular and/ or near circular symbols
that do not match with the optimum values of the pre set
reference parameters.
41/48
8. The system as claimed in claim 1, wherein the alphanumeric code
segregation module reads one or more alphanumeric codes from
within the one or more text data, by a customized OCR engine
which uses support vector machine classifier.
9. The system as claimed in claim 1, wherein the hyperlinking
modulereceives the one or more alphanumeric codes from the
alphanumeric code segregation module, further wherein the
hyperlinking module is configured to -
identify one or more workable input documents related with
each other, based on the one or more alphanumeric codes,
create one or more connected graphs between the one or
more related workable input documents; and
burn one or more clickable links onto the one or more
engineering drawing documents based on the one or more
connected graphs.
10. A method for hyperlinking one or more engineering drawing
documents, the method comprising the steps of -
42/48
acceptingthe one or more engineering drawing documents, and
generatingcorresponding one or more workable input
documents, the step comprising –
transformingthe one or more engineering drawing
documentsintoone or more system – ingestible formats,
distinctly, identifying, one or more images from the one or
more system – ingestible engineering drawing documents,
and
partitioning the one or more images into one or more image
segments, wherein the image segments are disjoint regions
bearing homogeneity in terms of properties including but not
limited to grey value or texture,
further wherein the combined proceeds of the sub – steps of
transforming, distinctly identifying, and partitioning, yields the
workable input documents,
locating the one or morecallout regions from within the one or
more workable input documents, the step comprising –
identifying one or more holes in the one or more workable
input documents,
identifying one or more circular symbolsand/ or near circular
symbols from within the one or more holes,
43/48
identifying one or more callout regions from within the one or
more circular symbols and/ or near circular symbols,
separating out one or more text data fromwithin the one or more
callout regions,
readingone or more alphanumeric codes from within the one or
more text data, wherein the alphanumeric codes include without
limitation, sheet number and/ or indexation information; and
establishing one or more hyperlinks between the one or more
engineering drawing documents, based on the one or more
alphanumeric codes.
11. The method as claimed in claim 11, wherein the step of accepting
the one one or more engineering drawing documents and
generating corresponding workable input documents, comprises
transforming the one or more engineering drawing documents into
one or more binary formatdocuments, the transforming being
achieved by -
converting the one or more engineering drawing documents into
corresponding one or more RGB format documents,
converting the one or more RGB format documents into
corresponding one or more greyscale format documents; and
44/48
converting the one or more greyscale format documents into
corresponding one or more binary format documents.
12. The method as claimed in claim 11, wherein the step of accepting
the one or more engineering drawing documents and generating
corresponding workable input documents, comprises partitioning
the one or more binary format documents into the one or more
image segments, the partitioning being achieved by assigning
different pixels to the background and foreground of the one or
more binary format documents, based on varying greyscale
intensity.
13. The method as claimed in claim 11, wherein the step of accepting
the one or more engineering drawing documents and generating
corresponding workable input documents, comprises partitioning
the one or more binary format documents into the one or more
image segments, the partitioning being achieved by watershed
technique and/or its variations.
45/48
14. The method as claimed in claim 11, wherein the step of locatingthe
one or more callout regions from within the one or more binary
format documents is achieved by -
smoothening the one or more images from within the binary
format documents, by applying one or more filters, including but
not limited to Gaussian filter,
obtaining one or more edge gradients and one or more pixel
directions corresponding to the one or more smoothened
images,
creating one or more edges based on the edge gradients and
pixel directions,
connecting the edges to obtain one or more preliminary edge
images,
removing any unwanted pixels that do not constitute the one or
more edges corresponding to the one or more edge images,
determining one or more final edge images from the preliminary
edge images by eliminating those edges which possess
improper edge gradient intensity and/ or which are unconnected;
and
identifying one or more hole bearing edge images from the one
or more final edge images by eliminating those one or more final
46/48
edge images that do not match one or more criteria
corresponding to circular shaped regions, the criteria including
height to width ratio and/ or black to white pixel ratio.
15. The method as claimed in claim 11, whereinthe step of locating the
one or more callout regions from the workable input documents,
comprises identifying the one or more circular and/ or near circular
symbols from the one or more circles by applying a modified Hough
Transformation technique that uses reference parameters including
without limitation, voting threshold for centre of circle, minimum
radius, maximum radius, and minimum distance between two
centers of circles.
16. The method as claimed in claim 11, wherein the step of locating the
one or more callout regions from the workable input documents,
comprises identifying one or more callout regions from within the
one or more circular symbols and/ or near circular symbols by –
pre setting the one or more reference parameters of the
circular and/ or near circular symbols to an optimum value,
the reference parameters including voting threshold for
centre of circle, minimum radius, maximum radius, and
minimum distance between two centers of circles; and
47/48
discarding the one or more circular and/ or near circular
symbols that do not match with the optimum values of the
pre set reference parameters.
17. The method as claimed in claim 11, wherein thestep of reading the
one or more alphanumeric codes from within the one or more text
data, is achieved by a customized OCR engine which uses support
vector machine classifier.
18. The method as claimed in claim 11, wherein the establishing of one
or more hyperlinks between the one or more engineering drawing
documents is achieved by -
identifyingone or more workable input documents related
with each other, based on the one or more alphanumeric
codes,
creating one or more connected graphs between the one or
more relatedworkable input documents; and
burningone or more clickable links onto the one or more
engineering drawing documents based on the one or more
connected graphs.