A method for removing similar targets by thermal imaging recognition

Abstract

The present invention relates to the field of thermal imaging technology, and particularly to a method for thermal imaging recognition and removal of similar targets. An embodiment of the present invention provides a method for thermal imaging recognition and removal of similar targets, including: pre-scanning all monitoring areas to form a preliminary target comparison image; wherein, a plurality of targets are included in the preliminary target comparison image, and the targets include detection threshold information; identifying similar targets in the preliminary target comparison image; collecting the detection threshold information of the similar targets to form a template matching library; monitoring the monitoring area in real time, and judging whether a target is a similar target according to a detection rule, a confidence value, and the template matching library; wherein, the detection rule and the confidence value are generated in a self-learning service. An embodiment of the present invention provides a method for thermal imaging recognition and removal of similar targets, which can remove similar targets.

Description

Technical Field

[0001] The present invention relates to the field of thermal imaging technology, and particularly to a method for removing similar targets in thermal imaging recognition. Background Art

[0002] With the rapid development of artificial intelligence technology, more and more industrial fields have added artificial intelligence target recognition technology, and thermal imaging video analysis and processing are no exception. Due to the characteristics of thermal imaging based on temperature imaging, it is determined that the amount of information of the target in the analysis is much less than that of visible light. If there are interference sources of similar targets in the visible area, target misdetection will occur.

[0003] Therefore, in view of the above deficiencies, there is an urgent need for a method for removing similar targets in thermal imaging recognition. Summary of the Invention

[0004] An embodiment of the present invention provides a method for removing similar targets in thermal imaging recognition, which can remove similar targets.

[0005] An embodiment of the present invention provides a method for removing similar targets in thermal imaging recognition, including:

[0006] Pre-scan all monitoring areas to form a preliminary target comparison image; wherein, the preliminary target comparison image includes multiple targets, and the targets include detection threshold information;

[0007] Identify similar targets in the preliminary target comparison image;

[0008] Collect the detection threshold information of the similar targets to form a template matching library;

[0009] Real-time monitor the monitoring area, and judge whether the target is the similar target according to the detection rule, confidence value and the template matching library; wherein, the detection rule and the confidence value are generated in the self-learning service.

[0010] In a possible design, the identifying similar targets in the preliminary target comparison image includes:

[0011] Mark the targets greater than the preset threshold;

[0012] Highlight the targets greater than the preset threshold;

[0013] Manually confirm whether the highlighted targets are similar targets.

[0014] In a possible design, the marking the targets greater than the preset threshold includes:

[0015] Mark the targets greater than the first preset threshold;

[0016] Mark the target greater than a first preset threshold; wherein, the first preset threshold is less than the second preset threshold;

[0017] The highlighting and identifying the target greater than the preset threshold includes:

[0018] Highlighting and identifying the target greater than the first preset threshold with a first color;

[0019] Highlighting and identifying the target greater than the second preset threshold with a second color.

[0020] In a possible design, the real-time monitoring of the monitoring area and determining whether the target is the similar target according to the detection rule, the confidence value, and the template matching library includes:

[0021] Real-time detecting the monitoring area and identifying the detection threshold for collecting the target;

[0022] If the target exists in the template matching library and is not the target to be blocked by the detection rule, and has exceeded the confidence value, then the target is a valid identified target; if the target does not exist in the template matching library and the target is within the detection rule, then the target is a valid identified target; otherwise, it is the similar target and is eliminated.

[0023] The present invention has at least the following beneficial effects compared with the prior art:

[0024] In this embodiment, by establishing a template matching library, collecting target information, generating a detection rule and a confidence value through a self-learning service to determine whether the target is a similar target, the problems of many similar targets and high false detection rate in thermal imaging target recognition are solved. The method provided by the present invention can effectively remove the interference source of similar targets that meet the removal conditions in the thermal imaging visible area. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0026] Figure 1 is a flowchart of a method for thermal imaging recognition and removal of similar targets provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0028] In the description of the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance; unless otherwise specified or stated, the term "plural" means two or more; the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, an integral connection, or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0029] In the description of this specification, it should be understood that the orientation terms such as "upper" and "lower" described in the embodiments of the present invention are described from the angles shown in the accompanying drawings and should not be construed as limiting the embodiments of the present invention. In addition, in the context, it should also be understood that when it is mentioned that an element is connected "above" or "below" another element, it can not only be directly connected "above" or "below" another element, but also be indirectly connected "above" or "below" another element through an intermediate element.

[0030] The embodiments of the present invention provide a method for removing similar targets by thermal imaging recognition, including:

[0031] Pre-scan all monitoring areas to form a preliminary target comparison image; among them, the preliminary target comparison image includes multiple targets, and the targets include detection threshold information;

[0032] Identify similar targets in the preliminary target comparison image;

[0033] Collect the detection threshold information of similar targets to form a template matching library;

[0034] Real-time monitor the monitoring area, and judge whether the target is a similar target according to the detection rule, confidence value, and template matching library; among them, the detection rule and confidence value are generated through the self-learning service.

[0035] In this embodiment, the thermal imaging video surveillance device is rotated to pre-scan all surveillance areas to form a preliminary target comparison image. By establishing a template matching library and collecting target information, detection rules and confidence values are generated through self-learning services to determine whether the target is a similar target, solving the problems of many similar targets and high false detection rate in thermal imaging target recognition. The method provided by the present invention can effectively remove similar target interference sources that meet the removal conditions in the thermal imaging visible area.

[0036] Specifically, the formation of the rule includes that in area detection, there are building targets similar to human figures in the surveillance screen. The surface temperature of the building changes throughout the year, and the significant target shape formed in the image also changes. However, through statistics, there is always a fitted change curve between this significant target shape and temperature. When performing target matching, this curve can be used as a rule to judge the target. If the detected target coincides with this time-shape curve, it is a similar target and no intelligent target analysis is required. The confidence value is the collected target threshold, and a model is obtained through deep learning to identify the targets in the image.

[0037] In some embodiments of the present invention, identifying similar targets in the preliminary target comparison image includes:

[0038] Marking targets greater than a preset threshold;

[0039] Highlighting the targets greater than the preset threshold;

[0040] Manually confirming whether the highlighted targets are similar targets.

[0041] In some embodiments of the present invention, marking targets greater than a preset threshold includes:

[0042] Marking targets greater than a first preset threshold;

[0043] Marking targets greater than a first preset threshold; wherein the first preset threshold is less than the second preset threshold;

[0044] Highlighting the targets greater than the preset threshold includes:

[0045] Highlighting the targets greater than the first preset threshold in a first color;

[0046] Highlighting the targets greater than the second preset threshold in a second color.

[0047] Acquire and identify the targets existing in the image and greater than the first detection threshold of 50, and mark them with a yellow box; acquire and identify the targets with a detection threshold greater than 80, and highlight them with a red box; the red box indicates that the customer needs to focus on identification and confirmation; the yellow box represents that the customer needs to identify and confirm. When placing the results after identification and confirmation into the template matching library, the current threshold of the target needs to be stored together in it.

[0048] In some embodiments of the present invention, the monitoring area is monitored in real time, and it is judged whether the target is a similar target according to the detection rules, confidence values, and template matching library, including:

[0049] Detect the monitoring area in real time and identify the detection threshold of the acquired target;

[0050] If the target exists in the template matching library and is not the target to be blocked by the detection rules, and has exceeded the confidence value, then the target is a valid identified target; if the target does not exist in the template matching library and the target is within the detection rules, then the target is a valid identified target; otherwise, it is a similar target and is excluded.

[0051] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for removing similar targets by thermal imaging recognition, characterized in that, Including: Performing a pre-scan on all monitored areas to form a preliminary target comparison image; wherein, the preliminary target comparison image includes multiple targets, and the targets include detection threshold information; Identifying similar targets in the preliminary target comparison image; Collecting the detection threshold information of the similar targets to form a template matching library; Real-time monitoring of the monitored area, and judging whether the target is the similar target according to the detection rule, the confidence value and the template matching library; wherein, the detection rule and the confidence value are generated in the self-learning service; The identifying similar targets in the preliminary target comparison image includes: Marking the targets greater than a preset threshold; Highlighting the targets greater than the preset threshold; Manually confirming whether the highlighted targets are similar targets; The marking the targets greater than a preset threshold includes: Marking the targets greater than a first preset threshold; Marking the targets greater than a second preset threshold; wherein, the first preset threshold is less than the second preset threshold; The highlighting the targets greater than the preset threshold includes: Highlighting the targets greater than the first preset threshold with a first color; Highlighting the targets greater than the second preset threshold with a second color; The real-time monitoring of the monitored area and judging whether the target is the similar target according to the detection rule, the confidence value and the template matching library includes: Real-time detecting the monitored area and identifying the detection threshold of the collected target; If the target exists in the template matching library and is not the target to be blocked by the detection rule, and has exceeded the confidence value, then the target is a valid recognition target; if the target does not exist in the template matching library and the target is within the detection rule, then the target is a valid recognition target; otherwise, it is the similar target and is excluded; The detection rule includes: fitting a time-profile curve based on the profiles of significant targets on the monitored screen of area detection at different times, and the target that coincides with the time-profile curve is the similar target; The confidence value is the collected target threshold, and a model is obtained through deep learning to identify the targets in the image.

Images