Method and system for automatically generating a map of an indoor space
Examples of generating a map for an indoor space are described. In one example, a raw digital representation of the indoor space is obtained and preprocessed by at least one of marking a region-of-interest (ROI) in the raw digital representation and classifying the indoor space into a space-category to obtain a preprocessed representation. Using the preprocessed representation, a skeletal map is generated including indicators to identify elements of the indoor space. Revision inputs are received to modify an erroneous indicator from amongst the multiple indicators. The erroneous indicator is revised to a modified indicator and the modified indicator and unmodified indicators are used to create a modified skeletal map which is usable for localizing a mobile device in the indoor space.
1 . A method of automatically generating a map of an indoor space, the method comprising:
obtaining a raw digital representation of the indoor space;
preprocessing the raw digital representation to obtain a preprocessed representation by:
marking a region-of-interest (ROI) in the raw digital representation, the marking includes:
identifying extraneous matter in the raw digital representation, the extraneous matter comprising matter not relevant to a floor plan of the indoor space;
discarding the extraneous matter to select a portion of the floor plan in the raw digital representation; and
identifying a space-category of the indoor space based on a layout of the indoor space and identification of entities of the indoor space from the raw digital representation to classify the indoor space into the space category;
generating, from the preprocessed representation, a skeletal map of the indoor space comprising a plurality of indicators each to identify a structural element of the indoor space;
overlaying the skeletal map having georeferenced plurality of indicators with a fingerprint map for generating a map for the indoor space, the fingerprint map comprising a correlation of sensor and signal information for a sequence of positions that constitute a navigation path traversed by a mobile device relative to the indoor space;
receiving revision inputs to correct an incorrectly identified indicator from amongst the plurality of indicators, wherein the indicator is revised to a modified indicator; and
creating a modified skeletal map of the indoor space comprising the modified indicator and unmodified indicators, wherein the modified skeletal map is usable with a localization application for localizing the mobile device in the indoor space.
2 . The method of claim 1 , executed by a trained machine-learning model.
3 . The method of claim 2 , wherein the trained machine-learning model is a trained deep-learning model.
4 . The method of claim 2 , wherein the modified indicator is fed to the trained machine-learning model as training data.
5 . The method of claim 1 , wherein the raw digital representation is one of a georeferenced vectored representation, a georeferenced rasterized representation, a non-georeferenced vectored representation, and non-a georeferenced rasterized representation.
6 . The method of claim 1 , wherein the preprocessing comprises converting the raw digital representation into a predefined standardized format.
7 . The method of claim 1 , wherein the preprocessing comprises performing at least one of resolution enhancement, noise reduction, and image augmentation on the raw digital representation.
8 . The method of claim 1 , wherein the generating comprises applying a feature detection technique, the feature detection technique comprising at least one of an edge detection technique and a corner detection technique.
9 . The method of claim 1 , wherein the skeletal map is in a rasterized representation.
10 . The method of claim 9 , wherein the creating comprises converting the rasterized representation into a vectored representation.
11 . A server computing device for automatically generating a map of an indoor space, the server computing device comprising:
a processor; and
a memory storing a set of instructions, the instructions executable in the processor to:
obtain a raw digital representation of the indoor space;
preprocess the raw digital representation to obtain a preprocessed representation by:
marking a region-of-interest (ROI) in the raw digital representation, the marking includes:
identifying extraneous matter in the raw digital representation, the extraneous matter comprising matter not relevant to a floor plan of the indoor space;
discarding the extraneous matter to select a portion of the floor plan in the raw digital representation; and
identifying into a space-category of the indoor space based on a layout of the indoor space and identification of entities of the indoor space from the raw digital representation to classify the indoor space into the space category;
generate, from the preprocessed representation, a skeletal map of the indoor space comprising a plurality of indicators each to identify a structural element of the indoor space;
overlay the skeletal map having georeferenced plurality of indicators with a fingerprint map for generating a map for the indoor space, the fingerprint map comprising a correlation of sensor and signal information for a sequence of positions that constitute a navigation path traversed by a mobile device relative to the indoor space;
receive revision inputs to correct an incorrectly identified indicator from amongst the plurality of indicators, wherein the indicator is revised to a modified indicator; and
create a modified skeletal map of the indoor space comprising the modified indicator and unmodified indicators, wherein the modified skeletal map is usable with a localization application for localizing a mobile device in the indoor space.
12 . The server computing device of claim 11 implementing a trained machine-learning model.
13 . The server computing device of claim 12 further comprising instructions to feed the modified indicator to the trained machine-learning model as training data.
14 . The server computing device of claim 12 , wherein the trained machine-learning model is a trained deep-learning model.
15 . The server computing device of claim 11 , wherein the raw digital representation is one of a georeferenced vectored representation, a georeferenced rasterized representation, a non-georeferenced vectored representation, and non-a georeferenced rasterized representation.
16 . The server computing device of claim 11 further comprising instructions to convert the raw digital representation into a pre-defined standardized format.
17 . The server computing device of claim 11 further comprising instructions to perform at least one of resolution enhancement, noise reduction, and image augmentation on the raw digital representation.
18 . The server computing device of claim 11 further comprising instructions to apply a feature detection technique, the feature detection technique comprising at least one of an edge detection technique and a corner detection technique.
19 . The server computing device of claim 11 , wherein the skeletal map is in a rasterized representation.
20 . The server computing device of claim 19 , further comprising instructions to convert the rasterized representation into a vectored representation.