T 1169/08 11-08-2011

1. The notice of appeal and the corresponding statement of grounds comply with the requirements of Articles 106 to 108 EPC and Rule 99 EPC. The appeal is, thus, admissible.

2. Added subject-matter

2.1 Independent claim 1 results, primarily, from a combination of original claim 1 with the further limitation regarding step (d) of original claim 1 according to which the most probable displacements are determined with a criterion based on low acceleration of the reading device, as recited in original claim 8. The reference to "edges of the bar code" has been deleted and replaced by a reference to "transitions from one colour to another of the lines of the bar code" as resulting from the passage on page 12, lines 17-20, of the published application. Step (c) in claim 1 has been specified, in line with the teaching of the description on page 4, lines 29-35 and, for the expression "coincidenses" (correctly spelled "coincidences"), on page 13, line 26. In this respect, the Board was satisfied that the intermediate generalisation resulting from the selection in feature (c) of the aspect relating to the detection of coincidences of the transitions, while omitting the process steps relating to the acquisition of the corresponding coordinates, was allowable. In the Board's view, the skilled person would have indeed recognised that alternative algorithms for this purpose could have been envisaged without the necessity to determine the actual transitions coordinates.

Similarly, independent claims 10 and 13 derive primarily from original claims 13 and 16, respectively, with the additional amendments discussed above, resulting from the explicit reference in current claims 10 and 13 to the method of claim 1.

2.2 Dependent claims 2 to 7 are based on original claims 2 to 7, respectively. New claim 8 results from a combination of original claims 9 and 11. New claim 9 derives from original claim 12. Original claims 14 and 15 constitute the basis for new dependent claims 11 and 12, respectively; the formulation of the claims having been rearranged for reasons of clarity.

3. Clarity - Support

3.1 In the Board's judgement the reference to "a criterion based on low acceleration of the reading device" in feature (d) of claim 1 meets the requirements of Article 84 EPC 1973 as to clarity. It was observed, in this respect, that the use of the relative term "low acceleration" indeed reflects the actual teaching of the present disclosure according to which movements associated with relatively high accelerations are discarded, thus, de facto privileging sequences of images associated with lower accelerations. The Board holds therefore that the skilled person would have indeed been able to understand, in the context of the present invention, the meaning of this term which use in independent claim 1 was thus fully supported by the present disclosure and acceptable (cf. Case law of the Boards of Appeal, 6th. edition; II.B, § 1.2.3, last two paragraphs).

The definition of error functions for the determination of the most probable sequence of displacements being explicitly presented as optional (cf. published application, page 5, line 3), the incorporation of this feature in step (d) of independent claim 1 was hence not required.

3.2 Moreover, by specifying in feature (d) of original claim 1 the nature of the criterion which, according to the description, actually permits to identify the most probable sequence of displacement, the claim has been amended so as to incorporate all essential features of the invention. The Board is hence convinced that the extent of protection conferred by the independent claims is adequately supported by the current description.

4. Sufficiency of disclosure - Content of the description

4.1 The Board concurs with the examining division in their finding that the sole example actually disclosed in the application is incomplete and that common general knowledge does not suffice to allow its full implementation by the skilled person (cf. point 4.2 below). The objection according to which an embodiment of the invention, as defined in a dependent claim, could not be carried out was, however, withdrawn as a consequence of the deletion of corresponding dependent claim 10. Moreover, since the application documents, considered as a whole, provide sufficient information allowing the reproduction of the claimed invention in its whole ambit, the Board holds that the requirements of sufficiency of disclosure are actually met (cf. point 4.3 below). Finally, the Board concludes that the absence of an example embodying the claimed invention does not constitute per se a reason for refusing the application, insofar as the requirements of sufficiency are met (cf. point 4.4 below).

4.2 The recursive formula:

ek = max[ek-1, (vk -vk-1)(ak - ak-1)],

which describes an example of an error function required to determine the most probable sequence of displacements, is not fully defined since the initial values of the parameters ek, vk and ak corresponding, respectively, to the values of the error function, velocity and acceleration for each image capture and intervening in said definition, have not been provided, thus making an iterative calculation impossible.

The board is not convinced that the skilled person would have been able, on the basis of common general knowledge, to complete said teaching and to select the initial values of the error function, as put forward by the appellant.

By setting the parameters e1, v1, a1 to zero as suggested in the statement of grounds of appeal, the iterative formula will lead to the result:

e2 = v2 . a2, which is also equivalent to

e2 = (v2)**(2) * 1/Deltat

since, according to the appellant: a2 = (v2 - v1)*1/Deltat (cf. applicant's letter dated 4 April 2007), wherein Deltat describes the period between two successive image captures.

In the Board's judgement, this selection of values for the initial parameters is, however, not appropriate since this choice would lead to a value of e2 which will generally exceed the value of the expression

(vk -vk-1)(ak - ak-1) and will thus propagate as the maximum value in all recursions of the error function, which corresponds precisely to the situation that the error function is deemed to prevent. As a matter of fact, this selection of the initial parameters would make the determination of the most probable movement actually impossible.

In a letter dated 4 April 2007, filed during the examination proceedings, the applicant referred to a further alternative as to the initial values of the variables intervening in the definition of the iterative error function. Assuming, in this case, that the first two image captures are identified by the indexes 0 and 1, the applicant was of the opinion that the skilled person would have, for instance, selected the following values or conditions for the parameters e, v and a:

e0 = e1 = 0;

v0 = v1;

a0 = a1 = 0.

The Board notes that this selection of initial values would, however, lead to the same result of the error function for all the conceivable displacements resulting from the second image capture (corresponding to index 1) since the value of e1 is predetermined. Furthermore, this selection would lead to the following expression of e2:

e2 = (v2 -v1)**(2) * 1/Deltat,

thus, possibly overestimating the error value for movements with low variation of the acceleration, with the damaging consequence that the value e2, then obtained, propagates in all recursions of the error function.

The previous analysis illustrates the fact that the selection of the initial values for the definition of the error function is not trivial and directly affects the behaviour of the error function and thus the reliability of the process of determining the most probable sequence of displacements. Although the selection of the initial values does not require being optimal, it should nevertheless permit to fulfil the claimed purpose.

During the oral proceedings before the Board, the appellant acknowledged that the values proposed so far for the initial parameters were not satisfactory. A third line of argumentation was then put forward. In the appellant's view, the skilled person would have solved this difficulty by simply starting the iterative process with the third image capture. In the appellant's view, even if this approach led to a plurality of conceivable displacements for the three first image captures, the requirement that the following captures fit to the previous ones associated with the fact that sequences associated to relatively high accelerations were discarded, would nevertheless make this solution clearly viable. It was underlined that this approach corresponded to the solution actually implemented in the systems proposed for sale.

However, nothing in the description hints at this solution which can also not be considered straightforward for the skilled person. The Board is also not convinced that the error function so defined would provide fully reliable results. In this context, it was stressed that, depending on the movement of the reading device when capturing the first images, a high value of the error function propagating to the successive measurements could also not be excluded. The appellant did not dispute this finding but objected that the existence of possible isolated failures did not prevent the method from giving general satisfaction. In case of an impossibility to discriminate between multiple conceivable displacements, the recording process simply needed to be repeated.

In the Board's view, the absence of information as to the actual implementation of the error function would have led the skilled person to consider various possibilities, whether in terms of possible initial values, as initially put forward by the appellant, or by starting calculation of the error function with the third image capture, as finally submitted during the oral proceedings before the Board. All these solutions share a common drawback: the behaviour of the error function is difficult to evaluate depending on the circumstances. It would indeed be strongly influenced by either the initial values of the parameters intervening in the elaboration of the error function or the movement of the reading device when capturing the first images. This sole finding would make a comparison extremely awkward: the reliability of a method being not only dependent on the actual movement of the reading device (i.e. of the hand of the operator), but also on the bar code itself (i.e. the sequence of lines) which is to be read. For these reasons, the Board holds that the implementation of the claimed method on the basis of the error function as defined in the original disclosure required a contribution from the skilled person which exceeded mere routine work and amounted thus to undue burden.

Confirmation for this position is seen in the fact that the appellant itself apparently had difficulties in reacting to this objection regarding the question of sufficiency of disclosure since three different solutions had been successively proposed as to what was considered to derive in a straightforward manner from the current description for the skilled person when implementing the error function.

4.3 Since the dependent claim which explicitly included the definition of the error function has been deleted, the issue to be decided under Article 83 EPC 1973 amounts to determine whether the application documents contained, despite the absence of a complete example actually embodying the invention, sufficient information in order for the skilled person to carry out the invention as defined in independent claims 1, 10 and 13.

The Board concurs with the examining division in their finding that the skilled person would have had no particular difficulties in reproducing steps (a), (b) and (c) of independent claim 1 underlying the decision in suit. The amendments which had been made to claim 1 do not affect the substance of these steps so that the same conclusion applies to steps (a) to (c) of amended claim 1. Similarly, the Board does not identify any reason preventing the skilled person from reproducing feature (e) of current claim 1 on the basis of the present application documents, assuming that he would have indeed been able to determine the most probable sequence of displacements for the sequence of images as recited in feature (d).

The question of sufficiency of description of the method of claim 1 hinges thus solely on the question whether the skilled person would have been able to carry out step (d) of claim 1. By specifying in this step that the determination of the most probable sequence of displacements is carried out with a criterion based on low acceleration of the reading device, the appellant restricted the claim to the actual teaching of the present disclosure. In this respect, the Board is convinced that the mere evocation of this criterion is, as such, sufficient and that the skilled person would have no difficulties to work out solutions relying on the determination of relatively low accelerations. In its simplest form, the skilled person would have possibly determined conceivable displacements by comparing measured accelerations with threshold values, previously obtained from empirical values or on the basis of experimentation he would have set up himself. In a more sophisticated variation of this basic process, the skilled person would have certainly considered comparing the various measured accelerations with each other so as to retain movements associated with the lowest accelerations. Although incomplete in itself, the example of the error function in the current description would have nevertheless constituted an incentive for the skilled person to develop alternative error functions which would have somehow reflected the acceleration of the reading device and thus constituted a basis for discriminating between various movements of the reading device.

For these reasons, the Board is convinced that the application contained sufficient information allowing the invention, as defined in independent claims 1, 10 and 13, to be performed over its whole scope.

4.4 According to Rule 27(1)(e) EPC 1973, the description shall "describe in detail at least one way of carrying out the invention claimed using examples where appropriate and referring to the drawings, if any". The present Board (in a different composition) already decided in a previous case that the purpose of the "examples" evoked in Rule 27(1)(e) EPC 1973 appears primarily to be to complete an otherwise incomplete teaching. As a consequence, the application cannot be refused under this provision if the description is considered to describe, despite the presence of erroneous drawings and the resulting lack of examples actually embodying the invention, "one way of carrying out the invention" (cf. point 3 in decision T 990/07, not published).

Applied to the present circumstances and to independent claims 1, 10 and 13, this principle implies that the incomplete example regarding the definition of the error function and the absence of any other example actually embodying the present invention do not constitute, as such, a bar to the grant of a patent on the basis of the current application. Since, moreover, the Board holds that the requirements of Article 83 EPC 1973 are met (cf. point 4.3 above) the condition that the description describes in detail one way of carrying out the invention, recited in Rule 27(1)(e) EPC 1973, is therefore met. To avoid any misleading later interpretation of the claims, the applicant shall, however, in the course of the ensuing examination proceedings, ensure that the description is adapted so as to establish without ambiguity that the example of the error function is incomplete and, therefore, does not as such embody the claimed invention.