Object detection is a challenging task in computer vision that involves predicting both the class and the location of objects in an image. Most existing methods rely on convolutional neural networks and hand-crafted modules, such as anchor boxes and non-maximum suppression. Recently, a novel end-to-end approach called DETR was proposed, which uses a transformer encoder-decoder structure to model object detection as a set prediction problem. However, DETR suffers from some limitations, such as poor performance on small objects and slow convergence speed. In this paper, we propose FF-DETR, a feature-fusion detection transformer that improves the performance and convergence speed of DETR-like models. FF-DETR introduces three feature fusion modules: (1) Contour Fusion FPN, which fuses multi-scale features using self-attention and deformable convolution; (2) Position-Content Query Fusion, which initializes the content query features by fusing the position query features and the encoder output features; and (3) Global Decoder Layer Fusion, which fuses the outputs of each decoder layer and updates the position query features iteratively. We conduct experiments on the COCO dataset and show that FF-DETR outperforms DETR and other variants in terms of accuracy and efficiency.
KEYWORDS: Target detection, Signal to noise ratio, Digital filtering, Image segmentation, Image filtering, Target recognition, Detection and tracking algorithms, Image processing, Image processing algorithms and systems, Signal detection
For target detection within a large-field cluttered background from a long distance, several difficulties, involving low
contrast between target and background, little occupancy, illumination ununiformity caused by vignetting of lens, and
system noise, make it a challenging problem. The existing approaches to dim target detection can be roughly divided into
two categories: detection before tracking (DBT) and tracking before detection (TBD). The DBT-based scheme has been
widely used in practical applications due to its simplicity, but it often requires working in the situation with a higher
signal-to-noise ratio (SNR). In contrast, the TBD-based methods can provide impressive detection results even in the
cases of very low SNR; unfortunately, the large memory requirement and high computational load prevents these
methods from real-time tasks. In this paper, we propose a new method for dim target detection. We address this problem
by combining the advantages of the DBT-based scheme in computational efficiency and of the TBD-based in detection
capability. Our method first predicts the local background, and then employs the energy accumulation and median filter
to remove background clutter. The dim target is finally located by double window filtering together with an improved
high order correlation which speeds up the convergence. The proposed method is implemented on a hardware platform
and performs suitably in outside experiments.
Real-time target detection against strong (bright) background under daytime is a challenging and leading edge subject, and also is a key technique for imaging tracking system. Strong background makes CCD image sensor work in critical saturation state, and imaging target contrast is very low. It's very difficult to accurately and stably track due to the complex characteristics of imaging target, such as strong clutter background, low contrast, and low signal to noise radio (SNR). So the key techniques for detecting and tracking target are eliminating the disturbance of diffuse reflection and beacon, synchronous detection, improving the performance of real-time image processing with high frame rate and high sampling rate.
A robust strategy for detecting and tracking day-time target was proposed in this paper. A series of efficient approaches ware presented to improve performance of detection and tracking in precision and stability, including strong background and noise suppression, image enhancement, adaptive thresholding, region merging based on morphology, recognition and tracking algorithm and so on.
In the end, we summarized and built the effictive flow for detecting and tracking target against strong background under daytime. The results of combining computer simulation with practical detection experiments show that the above-mentioned approaches are feasible and significant for real-time tracking system.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.