Claims:WE CLAIM:
1. A system for detecting presence of a desired object in camera frames, the system comprising a voice guided command centre (VGCC) in wireless communication with a communication device and a surveillance unit, the VGCC comprising:
a database for storing a plurality of camera frames received from each of a plurality of cameras of the surveillance unit, wherein each camera captures the camera frames of at least one location;
a clipping module configured for clipping at least one object from at least one camera frame and recording alphanumeric data associated with the clipped object;
a classification system configured for categorizing the clipped object to at least one class according to pre-defined rules;
a catalog module configured for formatting data of the clipped object to include a searchable catalog comprising synonyms, numerals, slang and phonetic data, the catalog being formatted as plain text;
a transceiver for receiving a voice search request of presence of the desired object in the camera frames over a communications network from the communication device operated by a user;
a voice to text/text to voice converter for converting the voice search request to plain text information; and
a processor configured for checking for presence of the desired object by matching at least in part the plain text information of the voice search request with the searchable catalog of the at least one clipped object, and transmitting a status signal based on matching to the communication device of the presence of the desired object in the camera frames over the communications network via the transceiver.
2. The system as claimed in claim 1, wherein the clipping module clips at least one object from at least one camera frame by comparing the camera frame with a reference frame, wherein the reference frame includes an image of the at least one location in absence of the at least one object.
3. The system as claimed in claim 1, wherein the clipping module includes a timer for recording time when the at least one object was captured in the camera frames by the plurality of cameras.
4. A method for detecting presence of at least one desired object in camera frames, the method is executed utilizing a voice guided command centre (VGCC), the method comprising the steps of:
receiving a plurality of camera frames at a database of the VGCC from each of a plurality of cameras, wherein each camera captures the camera frames of at least one location;
clipping at least one object from at least one camera frame and recording alphanumeric data associated with the clipped object;
categorizing the clipped object to at least one class according to pre-defined rules, wherein the at least one class is pre-assigned a name;
formatting data of the clipped object to include a searchable catalog comprising synonyms, numerals, slang and phonetic data, the catalog being formatted as plain text;
receiving a voice search request of presence of the desired object in the camera frames from a communication device operated by a user over a communications network;
converting the voice search request to plain text information;
checking for presence of the desired object by matching at least in part the plain text information of the voice search request with the searchable catalog of the at least one clipped object; and
transmitting a status signal of the presence of the desired object in the camera frames based on matching, over the communications network from a transceiver of the VGCC to the communication device.
5. The method as claimed in claim 4, wherein the step of clipping the object further comprising clipping the object to one or more sub-objects depending upon one or more pre-defined parameters such as size of the object, complexity of the object such as overlapping of at least two objects, presence of alphanumeric data, etc.
6. The method as claimed in claim 4, wherein the pre-defined rules pertain to shape of the object, facial recognition, motion of the object, etc.
7. The method as claimed in claim 4, wherein based on the pre-defined rule pertaining to shape of the object and facial recognition, a class processor of a classification system categorizes the clipped object to at least one class based on comparing the clipped object with at least one reference object belonging to the at least one class.
8. The method as claimed in claim 4, wherein the step of categorizing the clipped object to at least one class further includes categorizing the clipped object to at least one sub-class belonging to the at least one categorized class.
9. The method as claimed in claim 8, wherein in absence of the at least one sub-class at the VGCC, a class processor of the classification system transmits a sub-class request to one or more data sources for retrieving the at least one sub-class data, wherein the sub-class request includes a copy of the clipped object.
10. The method as claimed in claim 4, wherein the step of transmitting the status signal further comprising the step of:
transmitting a first query signal along with the status signal to the communication device in absence of the desired object in the camera frames, wherein the first query signal facilitates the user to modify the voice search request, thereby enhancing probability of detecting the presence of the desired object in the camera frames.
11. The method as claimed in claim 4, wherein the step of transmitting the status signal further comprising the step of:
transmitting a second query signal along with the status signal to the communication device, wherein the second query signal requests the user if any further search needs to be performed such as a narrow search or a broader search at the database of the VGCC.
, Description:FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENTS RULES, 2003
COMPLETE SPECIFICATION
[See section 10, Rule 13]

A SYSTEM AND METHOD FOR DETECTING PRESENCE OF A DESIRED OBJECT IN CAMERA FRAMES UTILIZING A VOICE GUIDED COMMAND CENTRE;

PRICEWATERHOUSECOOPERS PVT. LTD., A CORPORATION ORGANISED AND EXISTING UNDER THE LAWS OF INDIA, WHOSE ADDRESS IS 252, VEER SAVARKAR MARG, NEXT TO MAYOR'S BUNGALOW, SHIVAJI PARK, DADAR, MUMBAI, MAHARASHTRA 400028, INDIA

THE FOLLOWING SPECIFICATION PARTICULARLY DESCRIBES THE INVENTION AND THE MANNER IN WHICH IT IS TO BE PERFORMED.

FIELD OF THE INVENTION
[0001] The present invention generally relates to surveillance systems. More particularly, the present invention relates to a system and method for detecting presence of a desired object in camera frames, wherein the camera frames are received from one or more cameras of a surveillance unit.
BACKGROUND OF THE INVENTION
[0002] Generally, surveillance units, such as camera surveillance units, include a plurality of cameras. The cameras may be a closed circuit television (CCTV) camera or an Internet protocol (IP) camera. The cameras capture video images of the surveillance region(s).
[0003] A command centre of a surveillance unit is the brain behind all the operations. While, the command centre operators have direct access to the surveillance unit and can utilize the surveillance unit in various scenarios, field operators like beat marshals may not get easy access to surveillance unit. Field operators, often have only wireless sets over which they communicate with the command centre operators. At times, these field operators may require very basic information such as whether a car with a certain number crossed a particular camera on a specific time of a day.
[0004] Further, during certain situations, the command centre operators may not be available for answering a request received from the field operators. During these circumstances, the field operators find it difficult to track a certain object which may have passed the region of a specific camera of the surveillance unit. As a result, tracking process of various objects gets delayed.
BRIEF SUMMARY OF THE INVENTION
[0005] One or more embodiments of the present invention provide a system and method for detecting presence of a desired object in camera frames.
[0006] Accordingly, the invention provides a system for detecting presence of at least one desired object in camera frames. The system comprising a voice guided command centre (VGCC) and a communication device. The VGCC comprising: a database for storing a plurality of camera frames received from each of a plurality of cameras, wherein each camera captures the camera frames of at least one location; a clipping module configured for clipping at least one object from at least one camera frame and recording alphanumeric data associated with the clipped object; a classification system configured for categorizing the clipped object to at least one class according to pre-defined rules; a catalog module configured for formatting data of the clipped object to include a searchable catalog comprising synonyms, numerals, slang and phonetic data, the catalog being formatted as plain text; a transceiver for receiving a voice search request of presence of the desired object in the camera frames over a communications network from the communication device operated by a user; a voice to text/text to voice converter for converting the voice search request to plain text information; and a processor configured for checking for presence of the desired object by matching at least in part the plain text information of the voice search request with the searchable catalog of the at least one clipped object, and transmitting a status signal based on matching to the communication device of the presence of the desired object in the camera frames over the communications network via the transceiver.
[0007] In another aspect of the invention, a method for detecting presence of at least one desired object in camera frames is provided. The method includes the steps of: receiving a plurality of camera frames at a database of the voice guided command centre (VGCC) from each of a plurality of cameras, wherein each camera captures the camera frames of at least one location; clipping at least one object from at least one camera frame and recording alphanumeric data associated with the clipped object; categorizing the clipped object to at least one class according to pre-defined rules, wherein the at least one class is pre-assigned a name; formatting data of the clipped object to include a searchable catalog comprising synonyms, numerals, slang and phonetic data, the catalog being formatted as plain text; receiving a voice search request of presence of the desired object in the camera frames from a communication device operated by a user over a communications network; converting a voice search request to plain text information; checking for presence of the desired object by matching at least in part the plain text information of the voice search request with the searchable catalog of the at least one clipped object; and transmitting a status signal of the presence of the desired object in the camera frames based on matching, over the communications network from a transceiver of the VGCC to the communication device.
BRIEF DESCRIPTION OF THE DRAWINGS
[0008] Reference will be made to embodiments of the invention, examples of which may be illustrated in the accompanying figures. These figures are intended to be illustrative, not limiting. Although the invention is generally described in the context of these embodiments, it should be understood that it is not intended to limit the scope of the invention to these particular embodiments.
[0009] Figure 1 illustrates a system for detecting presence of at least one desired object in camera frames, according to an embodiment of the present invention.
[0010] Figure 2 illustrates an example of categorizing a clipped object to at least one class based on pre-defined rules utilizing a classification system that is suitable for use in system of figure 1, according to an embodiment of the present invention.
[0011] Figure 3 is a data structure diagram illustrating an example of formatting data of the clipped object to include a searchable catalog comprising synonyms, numerals, slang and phonetic data utilizing a catalog module that is suitable for use in system of figure 1, according to an embodiment of the present invention.
[0012] Figure 4 illustrates the system of figure 1 in communication with one or more data sources for retrieving data related to various fields according to an embodiment of the invention.
[0013] Figure 5 illustrates a flowchart of a method for detecting presence of a desired object in camera frames according to an embodiment of the invention.
[0014] Figure 6 illustrates a flowchart of a method for categorizing the clipped object to at least one sub-class in absence of the at least one sub-class at a voice guided command centre (VGCC), according to an embodiment of the invention.
DETAILED DESCRIPTION OF THE INVENTION
[0015] Figure 1 illustrates a system 100 for detecting presence of at least one desired object in camera frames, according to an embodiment of the present invention. The system 100 includes a communication device 110, a voice guided command centre (VGCC) 120 and a surveillance unit 150 having a plurality of cameras.
[0016] The communication device 110 including a communication module 112.
[0017] The VGCC 120 includes a database 122, a clipping module 124, a classification system 126 having at least one class memory 130, a catalog module 134, a transceiver 136, a voice to text/text to voice converter 138 and a processor 140.
[0018] The surveillance unit 150 is in communication with the VGCC 120. The VGCC 120 is in communication with the communication device 110.
[0019] Within the VGCC 120, the transceiver 136, the clipping module 124, the classification system 126, the catalog module 134, the voice to text/text to voice converter 138 and the processor 140 are in communication with the database 122. The classification system 126 is in communication with the catalog module 134. The transceiver 136 is in communication with the voice to text/text to voice converter 138. The voice to text/text to voice converter 138 is in communication with the processor 140.
[0020] With reference to figure 1 of an embodiment of the invention, the system 100 detects presence of a desired object in camera frames. The camera frames are received from each of a plurality of cameras of the surveillance unit 150. Each camera captures camera frames of at least one location. Therefore, field of view of each camera of the surveillance unit 150 is configured to capture camera frames of at least one location.
[0021] In accordance with an embodiment of the invention, the clipping module 124 of the system 100 is configured for clipping at least one object from at least one camera frame stored at the database 122 of the VGCC 120. The clipping module 124 clips the object from the camera frame based on comparing the camera frame with a reference frame, wherein the reference frame includes an image of the at least one location without the presence of at least one object. For instance, the clipping module 124 clips two objects from the camera frame based on comparing with the reference frame. The reference frame includes only the image of the location without the presence of the two objects.
[0022] With reference to figure 1 of an embodiment of the invention, the clipping module 124 includes a timer. The timer records the time when the at least one object was captured in the camera frames by the plurality of cameras.
[0023] Further, the clipping module 124 records the location in which the clipped object was captured by correlating with the source address of the camera which transmitted the camera frame, wherein the camera frame including the clipped object. For instance, if the clipping module 124 clips an object A from a camera frame 1, the clipping module 124 records the location in which the clipped object A was captured by correlating the source address of the camera which transmitted the camera frame 1. The field of view of each camera is configured to capture camera frames of the at least one location. Hence, each camera is identified based on the at least one location the camera is configured to capture.
[0024] In accordance with an embodiment of the invention, the clipping module 124 clips the at least one object into one or more sub-objects depending upon one or more pre-defined parameters. The one or more pre-defined parameters include size of the object, complexity of the object such as overlapping of at least two objects, presence of alphanumeric data, etc. For instance, if the clipped object appears such that there is an overlapping or merging of at least two objects, then the clipping module 124 clips the at least one clipped object into one or more sub-objects. Once, the object is clipped by the clipping module 124, the clipping module 124 identifies alphanumeric data associated with the clipped object. The clipped object along with the alphanumeric data is stored at the database 122 of the VGCC 120.
[0025] The classification system 126 of the system 100 is configured for categorizing the clipped object to at least one class memory 130 according to pre-defined rules. The classification system retrieves the clipped object from the database 122 of the VGCC 120. The pre-defined rules pertain to shape of the object, facial recognition, motion of the object, etc. The pre-defined rules are generated in a sequential order.
[0026] Figure 2 illustrates an example of categorizing the clipped object to at least one class according to pre-defined rules. For instance, if the pre-defined rule pertaining to shape of the object is set first, the class processor 140 categorizes the clipped object to at least one class based on the shape of the clipped object. The categorization process is illustrated below:
[0027] Firstly, the class processor 140 compares the clipped object with at least one reference object present in each class. Comparison of the clipped object with the at least one reference object is based on shape of the object. The class processor 140 of the classification system 100 compares the clipped object with the at least one reference object in each class. If the clipped object matches at least in part with the at least one reference object, the clipped object is categorized into the class wherein the at least one reference object is present. For instance, there are three different classes: vehicle class 1, human class 2 and animal class 3 as shown in figure 2.
[0028] Each of the classes as mentioned above will have at least one reference object. For instance, for the vehicle class 1, a car may be represented as a reference object. For the human class 2, a human may be represented as a reference object, and for the animal class, an animal may be represented as the reference object. The clipped object is compared with each of the reference objects present in each class respectively. Comparison is based on matching at least in part the clipped object with the reference objects present in each class. If the clipped object is identified as a vehicle based on comparison with the at least one reference object in each class, then the clipped object is categorized into the vehicle class 1 memory. The next step is categorizing the clipped object into one or more sub-classes. The sub-classes of the vehicle class 1 memory may include car sub-class memory 11, bike sub-class memory 12 and cycle sub-class memory 13. In view of the same, the clipped object currently categorized in vehicle class 1 memory is compared with at least one reference object present in each of the sub-classes, namely car sub-class memory 11, bike sub-class memory 12 and cycle sub-class memory 13. For instance, the clipped object is identified as a car based on the comparison, then the clipped object is categorized into car sub-class memory 11. Once the clipped object is categorized into car sub-class memory 11, next is to detect the type of car (clipped object). For detecting the type of car, the clipped object is compared with three sub-classes namely Mercedes sub-class memory 21, Honda sub-class memory 22 and Ford sub-class memory 23. Based on comparison of shape of the clipped object with shape of at least one reference object present in each sub-classes, the clipped object is identified as a Mercedes and the clipped object is assigned to the Mercedes sub-class memory 21. The example as illustrated is not limited to the class and/or sub-classes as mentioned in the figure 2. In view of the above illustrated example, the class processor 128 assigns a unique identifier to the clipped object such as clipped object: [vehicle, car, Mercedes]. The unique identifier representing the class and sub-classes of the clipped object.
[0029] In accordance with an embodiment of the invention, once the clipped object is categorized into class memory 130 and sub-classes, the catalog module 134 of the VGCC 120 formats data of the clipped object to include a searchable catalog comprising synonyms, numerals, slang and phonetic data, the catalog being formatted as plain text. The catalog module 134 accesses the data of the clipped object at the class memory 130 of the classification system 126. The catalog module 134 accesses the data of the clipped object by correlating with the unique identifier assigned to the clipped object. In the above illustrated example, the unique identifier for the clipped object is [vehicle, car, Mercedes]. The catalog module 134 formats the data of the mentioned clipped object to include synonyms, numerals, slang and phonetic data. With reference to figure 3, the data structure diagram illustrates an example of formatting data of the clipped object to include a searchable catalog comprising synonyms, numerals, slang and phonetic data utilizing a catalog module 134. According to the above example, for each of the identified class and sub-classes of the clipped object, namely vehicle, car and Mercedes, the corresponding synonyms, slang and phonetic data are generated as shown in figure 3.
[0030] In accordance with an embodiment of the invention, the alphanumeric data associated with the clipped object recorded at the clipping module 124, is also considered by the catalog module 134 as shown in figure 3.
[0031] The catalog module 134 includes an internal dictionary of synonyms, slang and phonetic data. Based on the categorized class and sub-classes, the catalog module 134 generates synonyms, slang and phonetic data. Further, the searchable catalog includes the alphanumeric data associated with the clipped object and location of the clipped object retrieved from the database 122 of the VGCC 120. For instance, if the clipped object is identified as the Mercedes car vehicle, then the alphanumeric data on name board of the clipped object will also form part of the searchable catalog. Further, the location in which the clipped object was captured will also form part of the clipped object data. The catalog module 134 formats the searchable catalog of the clipped object to plain text format. The searchable catalog of the clipped object is stored at the database 122 of the VGCC 120.
[0032] In accordance with an embodiment of the invention, a voice search request of presence of the desired object in the camera frames is received at the transceiver 136 of the VGCC 120 from the communication module 112 of the communication device 110. The communication device 110 is operated by a user such as a field operator, beat marshal, police operator, etc. For instance, the police operator sends a voice search request to the VGCC 120 to detect presence of a "Mercedes car vehicle with a name board of KA-1234 passed in a location X”. The police inspector communicates with the VGCC 120 by calling a designated number using the communication device 110. During the call, the police inspector sends the voice search request to the VGCC 120.
[0033] In accordance with an embodiment of the invention, the voice to text/text to voice converter 138 of the VGCC 120 converts the voice search request into plain text information. In the above mentioned example, the voice to text/text to voice converter 138 converts the voice search request "Mercedes car vehicle with a name board of KA-1234 passed in a location X” to plain text information and stores the plain text information of the voice search request at the database 122.
[0034] In accordance with an embodiment of the invention, the processor 140 of the VGCC 120 checks for presence of the desired object by matching at least in part the plain text information of the voice search request with the searchable catalog of the at least one clipped object data. For instance, the plain text information of the voice search request "Mercedes car vehicle with a name board of KA-1234 passed in a location X” is matched with the plain text of the searchable catalog of the clipped object data. As mentioned in the above example, the unique identifier assigned to the clipped object is [vehicle, car, Mercedes, KA-1234, location-X]. The processor 140 matches at least in part the plain text information “Mercedes car vehicle with a name board KA-1234 passed in location X” with plain text of the clipped object data, i.e. [vehicle, car, Mercedes, KA-1234, location-X]. The processor 140 of the VGCC 120 transmits a status signal of the presence of the desired object based on matching as mentioned above. The status signal indicates the presence of the desired object. For instance, the status signal indicates with a YES if the plain text information of the voice search request matches at least in part with plain text of the at least one clipped object data. Further, the status signal indicates with a NO if the plain text information of the voice search request does not match at least in part with the plain text of the at least one clipped object data. The processor 140 of the VGCC 120 transmits the status signal of the presence of the desired object from the VGCC 120 to the communication device 110 via the transceiver 136 of the VGCC 120. The status signal is transmitted either in a text format or in voice format depending upon requirements from the user of the communication device 110. The processor 140 of the VGCC 120 checks for presence of the desired object in real time and transmits the status signal also in real time.
[0035] In accordance with an embodiment of the invention, the processor 140 of the VGCC 120 transmits a first query signal along with the status signal, in absence of the desired object in the camera frames. The first query signal facilitates the user to modify the voice search request, thereby enhancing probability of detecting presence of the desired object in the camera frames. For instance, the status signal indicates to the user that the desired object “Mercedes car vehicle with a name board KA-1234 passed in location X” is not present in the camera frames, then the processor 140 transmits a first query signal to the user requesting the user to modify the voice search request. Based on the first query signal, the user can formulate alternative voice search requests.
[0036] In accordance with an embodiment of the invention, the processor 140 transmits a second query signal along with the status signal to the communication device 110. The second query signal requests the user if any further search needs to be performed such as a narrow search or a broader search at the database 122 of the VGCC 120. In response to the second query signal, the user can either narrow the voice search request or broaden the voice search request.
[0037] In accordance with an embodiment of the invention, if the user requests a narrow search at the VGCC 120, based on the narrow search, if the clipped object is not categorized into at least one sub-class, the class processor 140 assumes that the at least one sub-class is not available at the VGCC 120. During this situation, the class processor 140 retrieves the at least one sub-class data from one or more data sources as illustrated in figure 4 and stores at the classification system 126. The one or more data sources facilitate the system 100 in detecting at least one desired object. Firstly, a sub-class request is transmitted from the class processor 140 to the one or more data sources of presence of the at least one sub-class at the one or more data sources via the transceiver 136. For instance, the at least one clipped object is categorized into the human class. Subsequently, the class processor 140 makes an attempt to categorize the clipped object to human sub-class. If the human sub-class is not found, the class processor 140 sends a sub-class request to one or more data sources. The data sources are located external to the system 100 and do not form part of the system 100. The sub-class request will include a copy of the clipped object image. Based on the shape of the clipped object as illustrated in the above example, a copy of the clipped object will include a facial recognition of the clipped object, since the clipped object is categorized into the human class. The human sub-class is retrieved from the one or more data sources based on the facial recognition of the copy of the clipped object. The one or more data sources will transmit the human sub-class in response to receiving the facial image of clipped object. The human sub-class is stored at the classification system 126. The received human sub-class will include at least one image of the human similar to the clipped object. The human sub-class will include a name, as a result of which the clipped object human will automatically be assigned the name of the human sub-class. For instance, based on the facial recognition, the clipped object is assigned a name as Tom.
[0038] Further, the police inspector wishes to locate a human with the name Tom in a location. In view of the same, a voice search request “Is Tom present in location X” is transmitted to the VGCC 120 using the communication device 110. The voice search request is converted into plain text information by the voice to text/text to voice converter 138.
[0039] In accordance with an embodiment of the invention, the clipped object which is assigned the name Tom is formatted to include a searchable catalog. The searchable catalog comprises synonyms, numerals, slang and phonetic data. The searchable catalog being formatted as plain text.
[0040] The processor 140 of the VGCC 120 checks for presence of Tom at location X by matching at least in part the plain text of the clipped object assigned as Tom with plain text information of the voice search request “Tom”. A status signal is sent to the communication device 110 from the VGCC 120 of the presence of Tom in location X.
[0041] In accordance with an embodiment of the invention, the communication device 110, the VGCC 120 and the surveillance unit 150 of the system 100 communicate with each other over a communications network such as local area network (LAN), Metropolitan Area Network (MAN), Wide Area Network (WAN), Wireless Network and Inter Network, etc.
[0042] With reference to figure 5 of an embodiment of the invention, a flowchart of a method for detecting presence of at least one desired object in camera frames is illustrated. The method comprises the following steps:
[0043] At step 502, a plurality of camera frames is received at a database 122 of the VGCC 120 from each of a plurality of cameras of at least one location. Field of view of each camera of a surveillance unit is configured to capture camera frames of at least one location.
[0044] At step 504, at least one object is clipped from at least one camera frame using a clipping module of the VGCC. The clipping module clips the object from the camera frame based on comparing the camera frame with a reference frame. The reference frame includes an image of the at least one location without the presence of the at least one object. The time when the clipped object was captured is also recorded by a timer of the clipping module.
[0045] Further, any presence of alphanumeric data associated with the clipped object is recorded by the clipping module at step 506.
[0046] At step 508, the clipped object is categorized to at least one class according to pre-defined rules, wherein the at least one class is pre-assigned a name. A class processor of the classification system categorizes the clipped object to the at least one class. The pre-defined rules include shape of the object, facial recognition, motion of the object, etc. For instance, if the pre-defined rule pertains to shape of the object, the class processor will categorize the clipped object to at least one class based on the shape of the clipped object. Each class includes a reference object. The class processor compares clipped object with the at least one reference object in each class. If the clipped object matches at least in part with the at least one reference object, the clipped object is categorized into the class wherein the at least one reference object is present. Further, the clipped object is a unique identifier. The unique identifier will include information of the class in which the clipped object is categorized. For instance, the unique identifier of the clipped object will read as [vehicle, car, Mercedez]. The mentioned unique identifier indicates that the clipped object is a vehicle car of type Mercedez. The example of categorization is illustrated above with reference to figure 1 and figure 2 respectively. If the processor of the VGCC is unable to locate a sub-class for categorizing the clipped object, the processor assumes that the at least one sub-class is not available at the VGCC. During this situation, the class processor retrieves the at least one sub-class data from one or more data sources as illustrated in figure 6 and stores at the classification system. The one or more data sources facilitate the system 100 in detecting at least one desired object.
[0047] At step 510, data of the clipped object is formatted to include a searchable catalog comprising synonyms, numerals, slang and phonetic data. The searchable catalog is formatted as plain text. The catalog module accesses the clipped object data by correlating with the unique identifier of the clipped object. The catalog module formats the data of the mentioned clipped object to include synonyms, numerals, slang and phonetic data as illustrated in the example of figure 3. The catalog module includes an internal dictionary of synonyms, slang and phonetic data. Further, the alphanumeric data of the clipped object and the location where the clipped object was captured is also recorded at the catalog module.
[0048] At step 512, a voice search request of presence of the desired object in the camera frames is received at the VGCC from the communication device operated by a user. The voice search request is received over a communications network. Firstly, the user calls a designated number of the VGCC using a communication device, and during the call the voice search request is transmitted to the VGCC.
[0049] At step 514, the voice search request is converted to plain text information at a voice to text/text to voice converter. The plain text information is stored at the database of the VGCC.
[0050] At step 516, a processor of the VGCC checks for presence of the desired object by matching at least in part the plain text information of the voice search request with the searchable catalog of the at least one clipped object data. For instance, the plain text information of the voice search request "Mercedes car vehicle with a name board of KA-1234 passed in a location X” is matched with the plain text of the searchable catalog of the clipped object data. As mentioned in the above example, the unique identifier assigned to the clipped object is [vehicle, car, Mercedes, KA-1234, location-X]. The processor matches at least in part the plain text information “Mercedes car vehicle with a name board KA-1234 passed in location X” with plain text of the clipped object data, i.e. [vehicle, car, Mercedes, KA-1234, location-X].
[0051] At step 518, the processor of the VGCC transmits a status signal of the presence of the desired object based on matching in step 516. The status signal indicates the presence of the desired object. For instance, the status signal indicates with a YES if the plain text information of the voice search request matches at least in part with plain text of the at least one clipped object data. Further, the status signal indicates with a NO if the plain text information of the voice search request does not match at least in part with the plain text of the at least one clipped object data. The processor of the VGCC transmits the status signal of the presence of the desired object from the VGCC to the communication device via the transceiver of the VGCC. The status signal may be transmitted either in a text format or in voice format depending upon requirements from the user of the communication device. The processor of the VGCC checks for presence of the desired object in real time and transmits the status signal also in real time. Further, the processor of the VGCC transmits a first query signal along with the status signal, in absence of the desired object in the camera frames. The first query signal facilitates the user to modify the voice search request, thereby enhancing probability of detecting presence of the desired object in the camera frames. In yet another embodiment of the invention, if the user requests a narrow search at the VGCC, based on the narrow search if the clipped object is not categorized into at least one sub-class, the class processor assumes that the at least one sub-class is not available at the VGCC, and the class processor retrieves the at least one sub-class from one or more data sources as illustrated in figure 4. The one or more data sources facilitate the system in detecting at least one desired object. The example of retrieving sub-class information from the one or more data sources is indicated in the explanation with respect to figure 4 as mentioned above.
[0052] The foregoing description of the invention has been set merely to illustrate the invention and is not intended to be limiting. Since, modifications of the disclosed embodiments incorporating the spirit and substance of the invention may occur to person skilled in the art, the invention should be construed to include everything within the scope of appended claims.