Abstract:
In relation to a current image from a sequence of images captured by an image sensor, in a first step, a temporary motion vector is determined as a function of reference data comprising a preceding image and a motion vector associated to the preceding image. Then, in a second step, if the temporary motion vector does not satisfy a reliability criterion, the first step is repeated in relation to a following image, on the basis of the same reference data. Otherwise, the temporary motion vector is associated with the current image.
Abstract:
An image sequence sensor senses images. To associate a motion vector with an image of the sequence currently being processed, k candidate vectors are generated by adding, to a reference motion vector, respectively k search vectors. Then, a motion vector is selected from among the k candidate vectors as a function of a selection rule. Thereafter, the previous two steps are repeated m times, the reference motion vector being on the one hand, for a first iteration of the first step, an initial reference vector selected from among a set of vectors comprising at least one motion vector associated with a previous processed image and being on the other hand, for the m repetitions of the first step, the motion vector selected in the second step preceding the first step. Then, the vector obtained in the third step is associated with the image currently being processed.
Abstract:
A motion estimation method is provided for processing successive images in an image sequence, with a motion vector being associated with each of the processed images. For a current image, motion vectors associated with images that precede the current image in the sequence are selected. Candidate motion vectors are generated from the motion vectors that are selected. A motion vector is elected from among the candidate motion vectors. Information that associates the elected motion vector with the current image is stored in memory. At least one of candidate motion vectors is an acceleration vector generated from the acceleration between first and second motion vectors averaged relative to a first and second images, with the first and second images being distinct and preceding the current image in the image sequence. A motion vector averaged relative to a given image is obtained from selected motion vectors associated with images preceding the given image.
Abstract:
A method for converting an analog signal into a digital signal with automatic gain control includes inputting an analog signal into an analog-to-digital converter of a delta-sigma type having an output sampling frequency. The automatic gain control is performed in the analog-to-digital converter downstream of a delta-sigma modulator on an intermediate digital signal sampled at an intermediate sampling frequency greater than an output sampling frequency.
Abstract:
An integrated control and regulation circuit for a power stage of a regulated power supply, includes a current generator which, when the power supply is switched on, charges a decoupling capacitor to decouple a power stage of the power supply, through a first switch. The output from a logic circuit controls this first switch, and opens it when the regulated output voltage from the power stage reaches its nominal value. Preferably, a second switch controlled by the same output from the logic circuit deactivates a regulation loop of the power stage during the start up phase and in the case of a short circuit.
Abstract:
The invention provides a demodulation PLL wherein: the first position of a switch, which is controlled by a control circuit, respectively connects the outputs of a mixer and a LP filter to high gain and low gain inputs of an oscillator when frequency signals at the inputs of the mixer have not converged sufficiently, i.e. during the PLLs tuning mode; the second position of the switch respectively connects the outputs of the mixer and the LP filter to the low gain and high gain inputs of the oscillator when the frequency signals at the inputs of the mixer and the signal levels on the input and output of the filter have converged sufficiently, i.e. during the PLLs demodulation mode.