Abandoned Object Detection via CNN Image Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video surveillance solutions for detecting abandoned objects or illegal dumping are computationally expensive and require significant data processing, often compromising anonymity and efficiency.
Innovation Solution
A method utilizing a convolutional neural network (CNN) to analyze a subset of image data, blurring sensitive information, and implementing machine learning algorithms to detect and classify illegal dumping activities, reducing computational load and ensuring data anonymity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If video surveillance solutions process the complete stream of images to detect abandoned objects, then detection accuracy is improved, but computing time and data processing volume increase significantly
Solution Approach 1:
The patent divides the complete image stream into sampled subsets at reduced rates. Instead of analyzing every frame, the system selectively processes only certain frames or regions containing potential objects of interest, significantly reducing computational load while maintaining detection accuracy through strategic sampling.
Solution Approach 2:
The system performs partial processing by analyzing only a portion of the image data rather than the complete stream. By applying detection algorithms to sampled subsets and using change detection between samples, the system achieves sufficient detection accuracy without the excessive computational effort of processing all images.
2Reliability
If video surveillance solutions process large volumes of image data to detect illegal dumping, then detection reliability is improved, but data processing volume and computational resources increase significantly
Solution Approach 1:
The patent segments the large volume of image data into manageable sampled subsets. By processing only representative frames and using change detection algorithms comparing these samples, the system maintains reliable detection of illegal dumping activities while dramatically reducing the total data volume requiring computation.
Solution Approach 2:
The system extracts only the essential information needed for detection by comparing sampled frames and identifying changes. Instead of processing all image data, it extracts and analyzes only the differences between samples, which contain the critical information for detecting abandoned objects and illegal dumping.
3Measurement precision
If video surveillance solutions analyze complete image streams for abandoned objects, then detection precision is improved, but the complexity of the system increases
Solution Approach 1:
The patent reduces system complexity by segmenting the analysis task into processing discrete sampled frames rather than continuous streams. This approach simplifies the computational architecture while maintaining detection precision through strategic selection and comparison of sampled image subsets.
4Loss of information
If video surveillance solutions process all image data to ensure anonymity, then data protection is improved, but computational resources and processing time increase
Solution Approach 1:
The system applies anonymization processing only to the sampled subsets of images rather than the complete data stream. This partial processing approach maintains anonymity protection for the representative samples while avoiding the excessive processing time that would result from anonymizing every single frame.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
The invention relates to a computer-implemented method for detecting video sequences of interest of a zone under surveillance comprising: - acquiring a first video stream by means of a camera (10); - extracting a first series of images from the first video stream defining a second acquired image stream; - detecting a change specific to a deposit (20) within the second image stream; - extracting, by means of a component of the camera (10), a second series of images from the first acquired video stream at a second sampling frequency; - detecting a presence of a human (Ui) within at least one image of the second series of images; - transmitting an extract from the first video stream comprising the first date and the second date to a second remote server.