Software-only Implementation Research Articles

A Real-Time Implementation of Nonlinear Unsharp Masking with FPLDs

Real-time performance of image enhancement algorithms is required in many applications. In this paper we present a real-time image contrast enhancement scheme using non-linear unsharp masking, implemented with FPLDs (Field Programmable Logic Devices). A quadratic filter has been chosen for the nonlinear unsharp masking. The implementation performs well on real images from medical and non-medical fields. Performance figures of the nonlinear filter on mammograms show substantial improvements over software-only and other filter implementations using linear schemes, thus achieving real-time image enhancement of large images. The implementation has been tested as an application-specific co-processor to a standard PC, but can also be used as a functional unit of an application-specific image enhancement processor.

IMECO: A Reconfigurable FPGA-based Image Enhancement Co-Processor Framework

In this paper, we present a way to improve the computational speed of image contrast enhancement using low-cost FPGA-based hardware primarily targeted to X-ray images. In particular, we consider an enhancement method that consists of filtering followed by histogram modification. Filtering is done via the high boost filter (HBF) which is based on unsharp masking, and the histogram modification which is based on global histogram equalization (GHE). An image enhancement co-processor, IMECO, concept is proposed that enables efficient hardware implementation of enhancement procedures and hardware/software co-design to achieve high-performance low-cost solutions. The co-processor runs on an FPGA prototyping ISA-bus board. At this stage it consists of two hardware functional units that implement HBF and GHE and can be downloaded onto the board sequentially or reside on the board at the same time. These units represent an embryo of virtual hardware units that form a library of image enhancement algorithms. These algorithms can be easily integrated into software templates. In our trials with chest X-ray images, performance improvement over software-only implementations is more than two orders of magnitude, thus providing real-time or near-real-time image enhancement as required in target applications.

Design and applications of ATM LAN/WAN adapters

ATM networks are today operational, both as backbones for existing LAN technologies and as commercial wide-area multiservice networks. Still, in the early deployment of multi-site ATM networks, a number of difficulties have arisen out of the differences between the service definitions in local ATM networks and long-distance carrier networks. In particular, the adaptation of LAN emulation protocols relying on switched, best-effort connections to the first generation of ATM WAN services turned out to require the introduction of specific functions. In this paper, we study and discuss the nature of these adaptation functions which include peak rate shaping in order to comply with the traffic contract at the public UNI, efficient buffering and selective cell discarding to optimize the performance of end-to-end data protocols and fairness mechanisms to improve resource sharing. We describe a flexible hardware platform which enabled a quick prototyping of these functions. It provides on-board support for efficient cell processing and for associated functions (rate control, buffering, etc.). The use of a cell processor enables a software-only implementation of the ATM cell handling, yielding short development times and easy debugging while being compatible with an operation at line speeds up to 155 Mbit/s. We finally give examples and measurements of the use of the adapters in LAN/WAN interworking situations.

Scalable video transmission for the Internet

Information systems and distributed applications for the Internet show a growing demand for real-time audiovisual services. Support for those services within the Internet protocol layer will be widely available in the near future. Nevertheless, the heterogeneous structure of the Internet remains to be a great obstacle for establishing real-time video services. Scalable video codecs, generating bit-streams decodable at different rates, have been proposed to address the heterogeneity problem. In this article we present a scalable codec based on a spatio-temporal resolution pyramid combined with lattice vector quantization for efficient compression. With this approach additional lower bit-rate layers can be transmitted at the same overall bit-rate as needed by an H.263 single-layer transmission. The complexity of the codec is sufficiently low to allow software-only implementations of Internet video services. This is demonstrated on an Internet video server application.