Lossy Compression Algorithm Research Articles

NetBoost: Towards efficient distillation and service differentiation of network information exposure

Exposing network information such as distances between end hosts is useful to improve the quality of experiences for network users. Network providers calculate such information based on private topology and routing data and share it with users through well-established protocols such as Application-Layer Traffic Optimization. However, it is usually not intended to expose the original model, which can face scalability, user heterogeneity, efficacy & efficiency challenges.In this paper, we introduce NetBoost, a system that efficiently distills and differentiates ALTO-based network information to address these challenges concurrently. NetBoost significantly reduces the size of exposed network information by orders of magnitude. In particular, by utilizing a gradient boosting algorithm for classification and regression based on IP prefix matching, NetBoost provides high-order information exposure models, allowing network providers to offer differentiated services to clients with privacy-preserving. Our experimental results demonstrate that NetBoost performs effectively in resource-constrained scenarios, surpassing state-of-the-art lossy compression algorithms and achieving greater accuracy than the XGBoost gradient boosting algorithm, while maintaining a comparable compression rate. Furthermore, in simulation experiments conducted using the real-world networking software BIND, NetBoost achieved 6.96% and 5.2% higher accuracy compared to XGBoost under the same number of rules, with NetBoost’s accuracy set at 95% and 90%, respectively. Additionally, NetBoost reduced resolve time by 44.35% and 52.47%, respectively.

Compact lossy compression of tensors via neural tensor-train decomposition

AbstractMany real-world datasets are represented as tensors, i.e., multi-dimensional arrays of numerical values. Storing them without compression often requires substantial space, which grows exponentially with the order. While many tensor compression algorithms are available, many of them rely on strong data assumptions regarding its order, sparsity, rank, and smoothness. In this work, we propose TensorCodec, a lossy compression algorithm for general tensors that do not necessarily adhere to strong input data assumptions.TensorCodec incorporates three key ideas. The first idea is neural tensor-train decomposition (NTTD) where we integrate a recurrent neural network into Tensor-Train Decomposition to enhance its expressive power and alleviate the limitations imposed by the low-rank assumption. Another idea is to fold the input tensor into a higher-order tensor to reduce the space required by NTTD. Finally, the mode indices of the input tensor are reordered to reveal patterns that can be exploited by NTTD for improved approximation. In addition, we extend TensorCodec to enable the lossy compression of tensors with missing entries, often found in real-world datasets. Our analysis and experiments on 8 real-world datasets demonstrate that TensorCodec is (a) Concise: it gives up to $$7.38 \times $$ 7.38 × more compact compression than the best competitor with similar reconstruction error, (b) Accurate: given the same budget for compressed size, it yields up to $$3.33\times $$ 3.33 × more accurate reconstruction than the best competitor, (c) Scalable: Its empirical compression time is linear in the number of tensor entries, and it reconstructs each entry in logarithmic time. Our code and datasets are available at https://github.com/kbrother/TensorCodec.

Lossy Image Compression with Stochastic Quantization

Introduction. Lossy image compression algorithms play a crucial role in various domains, including graphics, and image processing. As image information density increases, so do the resources required for processing and transmission. One of the most prominent approaches to address this challenge is color quantization, proposed by Orchard et al. (1991). This technique optimally maps each pixel of an image to a color from a limited palette, maintaining image resolution while significantly reducing information content. Color quantization can be interpreted as a clustering problem (Krishna et al. (1997), Wan (2019)), where image pixels are represented in a three-dimensional space, with each axis corresponding to the intensity of an RGB channel. The purpose of the paper. Scaling of traditional algorithms like K-Means can be challenging for large data, such as modern images with millions of colors. This paper reframes color quantization as a three-dimensional stochastic transportation problem between the set of image pixels and an optimal color palette, where the number of colors is a predefined hyperparameter. We employ Stochastic Quantization (SQ) with a seeding technique proposed by Arthur et al. (2007) to enhance the scalability of color quantization. This method introduces a probabilistic element to the quantization process, potentially improving efficiency and adaptability to diverse image characteristics. Results. To demonstrate the efficiency of our approach, we present experimental results using images from the ImageNet dataset. These experiments illustrate the performance of our Stochastic Quantization method in terms of compression quality, computational efficiency, and scalability compared to traditional color quantization techniques. Conclusions. This study introduces a scalable algorithm for solving the color quantization problem without memory constraints, demonstrating its efficiency on a subset of images from the ImageNet dataset. The convergence speed of the algorithm can be further enhanced by modifying the update rule with alternative methods to Stochastic Gradient Descent (SGD) that incorporate adaptive learning rates. Moreover, the stochastic nature of the proposed solution enables the utilization of parallelization techniques to simultaneously update the positions of multiple quants, potentially leading to significant performance improvements. This aspect of parallelization and its impact on algorithm efficiency presents a topic for future research. The proposed method not only addresses the limitations of existing color quantization techniques but also opens up new possibilities for optimizing image compression algorithms in resource-constrained environments. Keywords: non-convex optimization, stochastic optimization, stochastic quantization, color quantization, lossy compression.

Massive Compression for High Data Rate Macromolecular Crystallography (HDRMX): Impact on Diffraction Data and Subsequent Structural Analysis.

New higher-count-rate, integrating, large area X-ray detectors with framing rates as high as 17,400 images per second are beginning to be available. These will soon be used for specialized MX experiments but will require optimal lossy compression algorithms to enable systems to keep up with data throughput. Some information may be lost. Can we minimize this loss with acceptable impact on structural information?

Electrocardiogram Signal Compression Using Deep Convolutional Autoencoder with Constant Error and Flexible Compression Rate

ObjectivesElectrocardiogram (ECG) signals are beneficial for diagnosing cardiac diseases. The cardiac patients' life quality likely increases with continuous or long-period recording and monitoring of ECG signals, leading to better and early diagnosis of disease and heart attacks. However, continuous ECG recording necessitates high data rates and storage, which means high costs. Therefore, ECG compression is a handy concept that facilitates continuous monitoring of ECG signals. Deep neural networks open up new horizons for compression and also for ECG compression by providing high compression rates and quality. Although they bring constant compression ratios with better average quality, the compression quality of individual samples is not guaranteed, which may lead to misdiagnoses. This study aims to investigate the effect of compression quality on the diagnoses and to develop a deep neural network-based compression strategy that guarantees a quality-bound in return for varying compression ratios.Materials and methodsThe effect of the compression quality on the arrhythmia diagnoses is tested by comparing the performance of the deep learning-based ECG classifier on the original ECG recordings and the distorted recordings using a lossy compression algorithm with different compression error levels. Then, a compression error upper limit is calculated in terms of normalized percent root mean square difference (PRDN) error, which also coincides with the findings of the previous studies in the literature. Lastly, to enable deep learning in ECG compression, a single encoder-multi-decoder convolutional autoencoder architecture, and multiple quantization levels are proposed to guarantee a desired upper limit on the error rate.ResultsThe efficiency of the proposed method is demonstrated on a popular benchmark data set for ECG compression methods using a transfer learning approach. The PRDN error is fixed to various values, and the average compression rates are reported. An average of 13.019:1 compression is achieved for a 10% PRDN error rate, assessed as a fair quality threshold for reconstruction error. It has also been shown that the compression model has a runtime that can be run in real-time on wearable devices such as commercial smartwatches.ConclusionThis study proposes a deep learning-based ECG compression algorithm that guarantees a desired upper limit on the compression error. This model may facilitate an eHealth solution for continuous monitoring of ECG signals of individuals, especially cardiac patients.

A lossy compression algorithm for data basis images based on DCT decomposition and diffuse representation

In this manuscript, we present a novel approach for compressing a database of images. Initially, we introduce a new representation that effectively reduces the data required to represent the image, enhancing the efficiency of subsequent compression steps. We apply the Discrete Cosine Transform (DCT) to the new representation and perform thresholding at different levels for a selected image. The DCT is known for concentrating image energy into a few coefficients, making it ideal for compression. By calculating the Peak Signal-to-Noise Ratio (PSNR) variation of the reconstructed image and evaluating the percentage of non-zero coefficients after thresholding, we plot the corresponding curve. These calculations provide insights into the quality of the reconstructed image, ensuring minimal degradation from compression. We repeat the steps with the Discrete Wavelet Transform (DWT), known for its multi-resolution analysis, to compare its performance against the DCT. The curves help us observe which process allows for image reconstruction with only a small percentage of coefficients, achieving significant data reduction. This comparative analysis determines the most effective method for various types of images. After analyzing the digital results from these curves, we generate and discuss the reconstructed image outcomes in detail. The digital and visual results presented at the conclusion demonstrate the robustness of the proposed strategy. We also explore potential applications in fields such as medical imaging, remote sensing, and multimedia storage, highlighting its versatility. Our findings suggest that the novel representation combined with DCT is highly effective, offering substantial storage space savings while maintaining image quality. These results are significant for theoretical research and practical implementations, providing a foundation for future advancements in image compression technologies.

How Much Storage Precision Can Be Lost: Guidance for Near-Lossless Compression of Untargeted Metabolomics Mass Spectrometry Data.

Several lossy compressors have achieved superior compression rates for mass spectrometry (MS) data at the cost of storage precision. Currently, the impacts of precision losses on MS data processing have not been thoroughly evaluated, which is critical for the future development of lossy compressors. We first evaluated different storage precision (32 bit and 64 bit) in lossless mzML files. We then applied 10 truncation transformations to generate precision-lossy files: five relative errors for intensities and five absolute errors for m/z values. MZmine3 and XCMS were used for feature detection and GNPS for compound annotation. Lastly, we compared Precision, Recall, F1 - score, and file sizes between lossy files and lossless files under different conditions. Overall, we revealed that the discrepancy between 32 and 64 bit precision was under 1%. We proposed an absolute m/z error of 10-4 and a relative intensity error of 2 × 10-2, adhering to a 5% error threshold (F1 - scores above 95%). For a stricter 1% error threshold (F1 - scores above 99%), an absolute m/z error of 2 × 10-5 and a relative intensity error of 2 × 10-3 were advised. This guidance aims to help researchers improve lossy compression algorithms and minimize the negative effects of precision losses on downstream data processing.

The Data Compression Method and FPGA Implementation in the Mars Rover Subsurface-Penetrating Radar on the Tianwen-1 Mission

Since Mars is far away from Earth, the propagation delay between Mars and Earth is very large. To ensure the effective use of the link transmission bandwidth, China’s first Mars exploration mission has put forward a demand for data compression for all scientific payloads. The on-board mature algorithms for data compression are mainly focused on optical images and microwave imaging radar applications. No articles have been published on data compression methods that are applied to subsurface-penetrating radar. Based on the background of this application, this paper proposes a logarithmic lossy compression algorithm which can meet the mission requirements for high compression ratios of 4:1 and 2.5:1. Its compression error is less than that of the block adaptive quantization (BAQ) algorithm. The algorithm is not only easy to implement on field-programmable gate array (FPGA) platforms, but also offers simple ground decompression and fast imaging. The experimental results show that high compression ratios of 4:1 and 2.5:1 can be realized, even if the data in and between traces do not have a strong correlation. And its relative error is less than 2%, which is a new type of high-efficiency data compression method that can be implemented on-board to meet with the demand of subsurface penetrating radar.

Perceptual Quality Improvement in Videoconferencing Using Keyframes-Based GAN

In the latest years, videoconferencing has taken a fundamental role in interpersonal relations, both for personal and business purposes. Lossy video compression algorithms are the enabling technology for videoconferencing, as they reduce the bandwidth required for real-time video streaming. However, lossy video compression decreases the perceived visual quality. Thus, many techniques for reducing compression artifacts and improving video visual quality have been proposed in recent years. In this work, we propose a novel GAN-based method for compression artifacts reduction in videoconferencing. Given that, in this context, the speaker is typically in front of the camera and remains the same for the entire duration of the transmission, we can maintain a set of reference keyframes of the person from the higher-quality I-frames that are transmitted within the video stream and exploit them to guide the visual quality improvement; a novel aspect of this approach is the update policy that maintains and updates a compact and effective set of reference keyframes. First, we extract multi-scale features from the compressed and reference frames. Then, our architecture combines these features in a progressive manner according to facial landmarks. This allows the restoration of the high-frequency details lost after the video compression. Experiments show that the proposed approach improves visual quality and generates photo-realistic results even with high compression rates. Code and pre-trained networks are publicly available at https://github.com/LorenzoAgnolucci/Keyframes-GAN.

MATHEMATICAL FRAMEWORK FORMULATION AND IMPLEMENTATION FOR HYPERSPECTRAL AEROSPACE IMAGES PROCESSING

This paper proposes a preprocessing algorithm for aerospace hyperspectral images based on a mathematical apparatus effectively applied in pre-compression transformation problems. In particular, several methods have been analyzed for hyperspectral image (signal) preprocessing from the point of view of digital signal processing algorithms. These mathematical methods are used for problems of filtering signals from noise of different natures and for compression and restoration of signals after their transmission through communication channels. The results of comparative analysis of preparatory processing of lossy compression algorithms based on wavelet analysis, discrete and orthogonal transforms are also given, demonstrating minimization of loss level of reconstructed decoded images. The performance of the proposed preprocessing algorithms with quality metrics is presented to evaluate the quality of the reconstructed hyperspectral aerospace images. The results of this study can be applied and used in the tasks of special processing of hyperspectral images, as well as fundamental knowledge of mathematical apparatuses of the proposed orthogonal preprocessing, considering the specificity of the data which is very important in obtaining images ready for compression for the subsequent identification of objects of the Earth's surface and using such mathematical transformations at the hyperspectral image preprocessing stage before compression provides efficient archiving of the obtained data, while reducing the communication channel load. Through the use of quality metrics of the reconstructed images, the preprocessing algorithm provides an understanding of the threshold of the peak signal-to-noise ratio value and the efficiency of its application to calculate and minimize the loss rate.

MIDOM—A DICOM-Based Medical Image Communication System

Despite the existing medical infrastructure being limited in terms of interoperability, the amount of medical multimedia transferred over the network and shared through various channels increases rapidly. In search of consultations with colleagues, medical professionals with the consent of their patients, usually exchange medical multimedia, mainly in the form of images, by using standard instant messaging services which utilize lossy compression algorithms. That consultation paradigm can easily lead to losses in image representation that can be misinterpreted and lead to the wrong diagnosis. This paper presents MIDOM—Medical Imaging and Diagnostics on the Move, a DICOM-based medical image communication system enhanced with a couple of variants of our previously developed custom lossless Classification and Blending Predictor Coder (CBPC) compression method. The system generally exploits the idea that end devices used by the general population and medical professionals alike are satisfactorily performant and energy-efficient, up to a point to support custom and complex compression methods successfully. The system has been implemented and appropriately integrated with Orthanc, a lightweight DICOM server, and a medical images storing PACS server. We benchmarked the system thoroughly with five real-world anonymized medical image sets in terms of compression ratios and latency reduction, aiming to simulate scenarios in which the availability of the medical services might be hardly reachable or in other ways limited. The results clearly show that our system enhanced with the compression methods in the question pays off in nearly every testing scenario by lowering the network latency to at least 60% of the latency required to send raw and uncompressed image sets and 25% in the best-case, while maintaining the perfect reconstruction of medical images and, thus, providing a more suitable environment for healthcare applications.

Impact of Image Compression on In Vitro Cell Migration Analysis

Collective cell movement is an indication of phenomena such as wound healing, embryonic morphogenesis, cancer invasion, and metastasis. Wound healing is a complicated cellular and biochemical procedure in which skin cells migrate from the wound boundaries into the wound area to reconstruct the injured skin layer(s). In vitro analysis of cell migration is an effective assay for measuring changes in cell migratory complement in response to experimental inspections. Open-source segmentation software (e.g., an ImageJ plug-in) is available to analyze images of in vitro scratch wound healing assays; however, often, these tools are error-prone when applied to, e.g., low-contrast, out-of-focus, and noisy images, and require manual tuning of various parameters, which is imprecise, tedious, and time-consuming. We propose two algorithmic methods (namely log gradient segmentation and entropy filter segmentation) for cell segmentation and the subsequent measurement of the collective cell migration in the corresponding microscopic imagery. We further investigate the effects of image compression on the algorithms’ measurement accuracy, applying lossy compression algorithms (the current ISO standards JPEG2000, JPEG, JPEG-XL and AV1, BPG, and WEBP). We aim to identify the most suitable compression algorithm that can be used for this purpose, relating rate–distortion performance as measured in terms of peak signal-to-noise ratio (PSNR) and the multiscale structural similarity index (MS-SSIM) to the segmentation accuracy obtained by the segmentation algorithms. The experimental results show that the log gradient segmentationalgorithm provides robust performance for segmenting the wound area, whereas the entropy filter segmentation algorithm is unstable for this purpose under certain circumstances. Additionally, the best-suited compression strategy is observed to be dependent on (i) the segmentation algorithm used and (ii) the actual data sequence being processed.

Sim-Piece: Highly Accurate Piecewise Linear Approximation through Similar Segment Merging

Approximating series of timestamped data points using a sequence of line segments with a maximum error guarantee is a fundamental data compression problem, termed as piecewise linear approximation (PLA). Due to the increasing need to analyze massive collections of time-series data in diverse domains, the problem has recently received significant attention, and recent PLA algorithms that have emerged do help us handle the overwhelming amount of information, at the cost of some precision loss. More specifically, these algorithms entail a trade-off between the maximum precision loss and the space savings achieved. However, advances in the area of lossless compression are undercutting the offerings of PLA techniques in real datasets. In this work, we propose Sim-Piece, a novel lossy compression algorithm for time-series data that optimizes the space requirements of representing PLA line segments, by finding the minimum number of groups we can organize these segments into, to represent them jointly. Our experimental evaluation demonstrates that our approach readily outperforms competing techniques, attaining compression ratios with more than twofold improvement on average over what PLA algorithms can offer. This allows for providing significantly higher accuracy with equivalent space requirements. Moreover, our algorithm, due to the simplicity of its merging phase, imposes little overhead while compacting the PLA description, offering a significantly improved trade-off between space and running time. The aforementioned benefits of our approach significantly improve the efficiency in which we can store time-series data, while allowing a tight maximum error in the representation of their values.

Optimization of Lossless Compression Algorithms using Multithreading

The process of reducing the number of bits required to characterize data is referred to as compression. The advantages of compression include a reduction in the time taken to transfer data from one point to another, and a reduction in the cost required for the storage space and network bandwidth. There are two types of compression algorithms namely lossy compression algorithm and lossless compression algorithm. Lossy algorithms find utility in compressing audio and video signals whereas lossless algorithms are used in compressing text messages. The advent of the internet and its worldwide usage has not only raised the utility but also the storage of text, audio and video files. These multimedia files demand more storage space as compared to traditional files. This has given rise to the requirement for an efficient compression algorithm. There is a considerable improvement in the computing performance of the machines due to the advent of the multi-core processor. However, this multi-core architecture is not used by compression algorithms. This paper shows the implementation of lossless compression algorithms namely the Lempel-Ziv-Markov Algorithm, BZip2 and ZLIB algorithms using the concept of multithreading. The results obtained prove that the ZLIB algorithm proves to be more efficient in terms of the time taken to compress and decompress the text. The comparison is done for both compressions without multithreading and compression with multi-threading.

Optimization of Lossless Compression Algorithms using Multithreading

The process of reducing the number of bits required to characterize data is referred to as compression. The advantages of compression include a reduction in the time taken to transfer data from one point to another, and a reduction in the cost required for the storage space and network bandwidth. There are two types of compression algorithms namely lossy compression algorithm and lossless compression algorithm. Lossy algorithms find utility in compressing audio and video signals whereas lossless algorithms are used in compressing text messages. The advent of the internet and its worldwide usage has not only raised the utility but also the storage of text, audio and video files. These multimedia files demand more storage space as compared to traditional files. This has given rise to the requirement for an efficient compression algorithm. There is a considerable improvement in the computing performance of the machines due to the advent of the multi-core processor. However, this multi-core architecture is not used by compression algorithms. This paper shows the implementation of lossless compression algorithms namely the Lempel-Ziv-Markov Algorithm, BZip2 and ZLIB algorithms using the concept of multithreading. The results obtained prove that the ZLIB algorithm proves to be more efficient in terms of the time taken to compress and decompress the text. The comparison is done for both compressions without multithreading and compression with multi-threading.

Evaluation of lossy compression algorithms using discrete cosine transform for sounding rocket vibration data

This work evaluates the performance of compression algorithms based on discrete cosine transform (DCT) when used to encode sounding rockets’ vibration signals. Different compression methods that use the three-step strategy (transform, quantization, and entropy encoder) are presented. The performances are evaluated using the relationship between the compression rate and distortion, the latter being measured using the peak signal-to-noise ratio (PSNR) and the difference observed in the power spectral density. The focus of the work is on vibration signals from rocket launches conducted by the Institute of Aeronautics and Space of the Brazilian Air Force. The obtained results present a great variation for the vibration data in the test set. Compressors operating with PSNR close to 50 dB can achieve very high compression rates (over 100) for low power signals. However, if the vibration signal strength increases significantly, the compression rate can be much lower. For the worst case in the test set (signal with the highest power), the work shows that the best proposed method needs a rate of 9.8 to achieve a PSNR of 48.2 dB. When compared to the performance of recently published DCT-based compression algorithms, the proposed technique shows significant improvements. Given a PSNR around 50 dB, for signal with moderate power, improvements above 100% are observed in compression rate.

BLASTNet: A call for community-involved big data in combustion machine learning

Many state-of-the-art machine learning (ML) fields rely on large datasets and massive deep learning models (with O(109) trainable parameters) to predict target variables accurately without overfitting. Within combustion, a wealth of data exists in the form of high-fidelity simulation data and detailed measurements that have been accumulating since the past decade. Yet, this data remains distributed and can be difficult to access. In this work, we present a realistic and feasible framework which combines (i) community involvement, (ii) public data repositories, and (iii) lossy compression algorithms for enabling broad access to high-fidelity data via a network-of-datasets approach. This Bearable Large Accessible Scientific Training Network-of-Datasets (BLASTNet) is consolidated on a community-hosted web-platform (at https://blastnet.github.io/), and is targeted towards improving accessibility to diverse scientific data for deep learning algorithms. For datasets that exceed the storage limitations in public ML repositories, we propose employing lossy compression algorithms on high-fidelity data, at the cost of introducing controllable amounts of error to the data. This framework leverages the well-known robustness of modern deep learning methods to noisy data, which we demonstrate is also applicable in combustion by training deep learning models on lossy direct numerical simulation (DNS) data in two completely different ML problems — one in combustion regime classification and the other in filtered reaction rate regression. Our results show that combustion DNS data can be compressed by at least 10-fold without affecting deep learning models, and that the resulting lossy errors can even improve their training. We thus call on the research community to help contribute to opening a bearable pathway towards accessible big data in combustion.

A lightweight, user-configurable detector ASIC digital architecture with on-chip data compression for MHz X-ray coherent diffraction imaging

Today, most X-ray pixel detectors used at light sources transmit raw pixel data off the detector ASIC. With the availability of more advanced ASIC technology nodes for scientific application, more digital functionalities from the computing domains (e.g., compression) can be integrated directly into a detector ASIC to increase data velocity. In this paper, we describe a lightweight, user-configurable detector ASIC digital architecture with on-chip compression which can be implemented in 130 nm technologies in a reasonable area on the ASIC periphery. In addition, we present a design to efficiently handle the variable data from the stream of parallel compressors. The architecture includes user-selectable lossy and lossless compression blocks. The impact of lossy compression algorithms is evaluated on simulated and experimental X-ray ptychography datasets. This architecture is a practical approach to increase pixel detector frame rates towards the continuous 1 MHz regime for not only coherent imaging techniques such as ptychography, but also for other diffraction techniques at X-ray light sources.

A deep learning scheme for efficient multimedia IoT data compression

Multimedia Internet of Things (MIoT) devices and networks will face many power and communication overhead constraints given the volume of multimedia sensed data. One classic approach to overcoming the difficulty of large-scale data is to use lossy compression. However, current lossy compression algorithms require a limited compression rate to maintain acceptable perceived image quality. This is commonly referred to as the image quality-compression ratio trade-off. Motivated by current breakthroughs in computer vision, this article proposes recovering high-quality decompressed images at the application server level using a deep learning-based super-resolution model. As a result, this paper proposes ignoring the trade-off between image quality and size and increasing the reduction size further by using a lossy compressor with downscaling to conserve energy. The experimental study demonstrates that the proposed technique effectively improves the visual quality of compressed and downscaled images. The proposed solution was evaluated on resource-constrained microcontrollers. The obtained results show that the transmission latency and energy consumption can be decreased by up to 10% compared to conventional lossy compression techniques.

A new fingerprint definition for effective song recognition

Music and song recognition is an activity of wide interest for researchers and companies due to the intrinsic challenges and the possible economic profits it can give. Despite basic algorithms about song recognition are simple in principle, it is quite difficult to obtain an efficient and robust approach able to generate an effective algorithm for identifying songs on the fly. This statement is proved by the fact that there are very few companies in the world having their core business into this field, even if the potential market is very huge. In this paper, we propose a new approach for generating fingerprints from excerpts of songs that is the first step in implementing a complete algorithm of song recognition. Their generation is based on the Welch’s method for spectral density estimation, the use of a Mel filter bank and an exponential adaptive threshold curve in the frequency domain never used before. Even if the previous techniques are not new, at the best of our knowledge they are not used all together for fingerprint generation. Our main purpose is to show that the proposed fingerprint generation approach permits to obtain a very high accuracy in recognizing pieces of song and their position inside the song, as well as it appears robust compared to typical alteration of the audio signal. Specifically, the fingerprints we generate are highly insensitive to noise and audio lossy compression algorithms; moreover, we think the method is prone also to generate pitch insensitive fingerprints with a small modification. We show through an experimentation with a large database of songs the recognition accuracy obtained with our fingerprints is better than the landmark-based approach (already used by the famous Shazam application). This is not a negligible results because even small improvements means a very large number of more recognitions, with higher profit prospects in industrial applications. In order to better focus on the fingerprint structure and its generation algorithm, we don’t discuss any specific search algorithm, that is a subject of further work, and we use a linear search only in our experiments; in such a way, we think the goodness of the fingerprint as such is better evinced.