Defendant, Theodore Alphonso Osborne, was convicted by a jury of robbery with a dangerous weapon. From a judgment imposing a thirty-five-year sentence for the Class D felony, defendant appeals. Defendant’s eight assignments of error relate to the admission of certain real and testimonial evidence, the trial court’s denial of defendant’s motion for a mistrial and motion to dismiss the charges at the conclusion of the evidence, and the court’s findings on mitigating and aggravating factors. We find no error.
I
The State’s evidence tended to show that on the morning of 10 November 1985 at approximately 12:50 a.m., an armed black man robbed the Fast Fare convenience store in Ahoskie, North Carolina, where Irene Bedgood was working as cashier. Holding Ms. Bedgood at gunpoint, the assailant demanded and took approximately $100 from the cash register as well as Ms. Bedgood’s pocketbook containing $110 to $120. He then backed out the door and fled.
*500Following the robbery, Ms. Bedgood accompanied the investigating officer to the police station where she assisted another officer in preparing a composite photograph of the robber. She helped prepare a second composite the following day. Two or three days later, Ms. Bedgood positively identified the defendant as the robber from a lineup of seven photographs.
At trial, the State’s primary inculpatory evidence consisted of Ms. Bedgood’s testimony, including her in-court identification of the defendant, and a pair of gloves, found near Ms. Bedgood’s recovered pocketbook and which matched gloves the defendant had been seen wearing at work. The defendant presented evidence of an alibi.
II
Defendant contends that Ms. Bedgood’s in-court identification of the defendant was tainted by an unnecessarily suggestive pretrial identification procedure and thus should not have been allowed. Prior to the presentation of any evidence to the jury, a voir dire was conducted concerning the in-court and out-of-court identification of the defendant. Based upon testimony of Ms. Bed-good and of Doug Doughtie, the investigating officer, and upon his own observation of the two composites and the seven photographs used in the lineup, the trial judge made numerous findings of fact and concluded that the pretrial identification procedure was not overly suggestive. He further determined that, even in the event of an unnecessarily suggestive photographic lineup, the in-court identification was of independent origin based upon Ms. Bedgood’s direct observations during the robbery.
[1] The test for determining whether an identification procedure violates due process is whether the procedure “in the totality of the circumstances ... is so unnecessarily suggestive and conducive to irreparable misidentification that it offends fundamental standards of decency and justice.” State v. Freeman, 313 N.C. 539, 544, 330 S.E. 2d 465, 471 (1985); State v. Grimes, 309 N.C. 606, 609, 308 S.E. 2d 293, 294 (1983). Defendant’s chief contention is that the preparation of a composite prior to the photographic lineup necessarily made the lineup suggestive. We disagree. Law enforcement officers often make use of composites when compiling live or photographic lineups, see, e.g., State v. Williams, 69 N.C. App. 126, 316 S.E. 2d 322 (1984), and there is nothing inherently *501unfair in this procedure nor a resulting lineup in which the defendant, of all the subjects, most closely resembles the composite. See State v. Montgomery, 291 N.C. 91, 229 S.E. 2d 572 (1976).
[2] The description given by Ms. Bedgood and the composite prepared with her assistance indicated that the perpetrator had closely cropped hair and slanted eyes. Due process does not require that all subjects in a photographic lineup be identical in appearance. See, e.g., Montgomery, 291 N.C. at 100, 229 S.E. 2d at 579. Nor is such a lineup impermissibly suggestive merely because the defendant has a distinctive appearance. Freeman, 313 N.C. at 545, 330 S.E. 2d at 471. All that is required is that the lineup be fair and that the investigating officers do nothing to induce the witness to select one subject rather than another. Id.; Montgomery, 291 N.C. at 100, 229 S.E. 2d at 579.
[3] We do not have the photographs before us on this appeal and must therefore rely upon Officer Doughtie’s testimony regarding their appearance. His testimony indicates that the seven photographs were of black males of varying size and build, hair length and degree of facial hair. Three photographs showed individuals with very short hair and three were of subjects with slanted eyes. The trial judge made findings that only two or three pictures were of individuals with physical characteristics substantially dissimilar to the defendant, and even as to those there were certain similarities. Furthermore, there is no evidence of any improper inducement of Ms. Bedgood to choose one subject over another. We conclude that the evidence supports the court’s findings and conclusion that the pretrial identification procedure, including the use of the composites and photographic lineup, was not impermissibly suggestive.
[4] Even when a pretrial identification procedure is invalid, an in-court identification is admissible if found to be of independent origin. State v. Clark, 301 N.C. 176, 270 S.E. 2d 425 (1980). Defendant argues that “conflicting evidence” given by Ms. Bedgood regarding her identification establishes that the in-court identification was not of independent origin. The witness testified that the man who robbed her was clean shaven except for a slight mustache. However, the two composites prepared almost immediately after the event showed no mustache or facial hair. On the other hand, the photograph of defendant chosen from the lineup *502showed him as having a mustache, a goatee and sideburns. The defendant testified that at the time of the robbery, he wore thick sideburns, a goatee, and a thick mustache.
Obviously a man may grow or shave his facial hair at will. And any perceived discrepancies or equivocation in Ms. Bedgood’s identification go merely to the weight of her testimony, not its competency. See State v. Pridgen, 313 N.C. 80, 326 S.E. 2d 618 (1985). The factors which determine whether an in-court identification is of independent origin are:
(1) the opportunity of the witness to view the criminal at the time of the crime; (2) the witness’s degree of attention; (3) the accuracy of the witness’s prior description of the criminal; (4) the level of certainty demonstrated by the witness at the confrontation; and (5) the length of time between the crime and the confrontation.
Freeman, 313 N.C. at 544, 330 S.E. 2d at 471; State v. Baldwin, 59 N.C. App. 430, 433, 297 S.E. 2d 188, 191 (1982), cert. denied, 307 N.C. 698, 301 S.E. 2d 390 (1983).
The testimony of Ms. Bedgood on voir dire and during the trial indicate that she observed the defendant attentively, face to face, at close range, in a well-lighted area, without obstruction, for three to four minutes during the course of the robbery. She accurately described the defendant immediately after the robbery and at trial. She positively identified the defendant from the lineup within two or three days of the crime and again at trial less than four months later. Moreover, Ms. Bedgood testified that her identification was based on her recollection of the robbery. We therefore hold that the Court’s determination that Ms. Bedgood’s in-court identification was of independent origin is supported, as it must be, by clear and convincing evidence. See State v. Yancey, 291 N.C. 656, 231 S.E. 2d 637 (1977).
Ill
We have carefully reviewed the defendant’s remaining seven assignments of error and find them to be without merit. Thus we conclude that defendant received a fair trial free of prejudicial error and that the sentence imposed was proper.
*503No error.
Judges WELLS and Orr concur.